MicroStrategy Functions Reference

Fu n ct i on s
Re f e r e n ce
Version 2021
M i cr o St r at egy 2021
Oct o b er 2023
Copyright © 2023 by MicroStrategy Incorporated. All rights reserved.
Trademark Information
The following are either trademarks or registered trademarks of MicroStrategy Incorporated or its affiliates in the United States and certain other
countries:
Dossier, Enterprise Semantic Graph, Expert.Now, Hyper.Now, HyperIntelligence, HyperMobile, HyperVision, HyperWeb, Intelligent Enterprise,
MicroStrategy, MicroStrategy 2019, MicroStrategy 2020, MicroStrategy 2021, MicroStrategy Analyst Pass, MicroStrategy Architect, MicroStrategy
Architect Pass, MicroStrategy Badge, MicroStrategy Cloud, MicroStrategy Cloud Intelligence, MicroStrategy Command Manager, MicroStrategy
Communicator, MicroStrategy Consulting, MicroStrategy Desktop, MicroStrategy Developer, MicroStrategy Distribution Services, MicroStrategy
Education, MicroStrategy Embedded Intelligence, MicroStrategy Enterprise Manager, MicroStrategy Federated Analytics, MicroStrategy Geospatial
Services, MicroStrategy Identity, MicroStrategy Identity Manager, MicroStrategy Identity Server, MicroStrategy Insights, MicroStrategy Integrity
Manager, MicroStrategy Intelligence Server, MicroStrategy Library, MicroStrategy Mobile, MicroStrategy Narrowcast Server, MicroStrategy ONE,
MicroStrategy Object Manager, MicroStrategy Office, MicroStrategy OLAP Services, MicroStrategy Parallel Relational In-Memory Engine (MicroStrategy
PRIME), MicroStrategy R Integration, MicroStrategy Report Services, MicroStrategy SDK, MicroStrategy System Manager, MicroStrategy Transaction
Services, MicroStrategy Usher, MicroStrategy Web, MicroStrategy Workstation, MicroStrategy World, Usher, and Zero-Click Intelligence.
The following design marks are either trademarks or registered trademarks of MicroStrategy Incorporated or its affiliates in the United States and certain
other countries:
Other product and company names mentioned herein may be the trademarks of their respective owners.
Specifications subject to change without notice. MicroStrategy is not responsible for errors or omissions. MicroStrategy makes no warranties or
commitments concerning the availability of future products or versions that may be planned or under development.
CON TEN TS
1. Understanding Functions in MicroStrategy 5
The Basics of MicroStrategy Objects, Object Definitions, and Functions 6

Function Syntax and Formula Components 9
Using Functions in Expressions 35
How MicroStrategy Processes Functions 60
Additional Examples of Functions in Expressions 83
2. Standard Functions 114
Basic Functions 116

Date and Time Functions 149
Internal Functions 169
NULL/Zero Functions 188
OLAP Functions 191
Pattern Functions 265
Rank and NTile Functions 268
String Functions 293
3. Operators 319
Arithmetic Operators 320

Comparison Operators 323
Comparison for Rank Operators 334
Logical Operators 338
4. Plug-In Package Functions 342
Data Mining Functions 343

Financial Functions 347
Mathematical Functions 415
Statistical Functions 439
A. MicroStrategy and Database Support for Functions 494
Analytical Engine Support for Functions 495

Databases That a Function Can Be Evaluated On 500
Fu n ct io n s Refer en ce
U N D ERSTAN D IN G
FUN CTION S IN
M ICRO STRATEGY
Copyright © 2023 All Rights Reserved 5

This chapter provides the following information:
l The Basics of MicroStrategy Objects, Object Definitions, and Functions,

page 6
l Function Syntax and Formula Components, page 9
l Using Functions in Expressions, page 35
l Adding Functions to Expressions with the Insert Function Wizard, page 57
l How MicroStrategy Processes Functions, page 60
l Additional Examples of Functions in Expressions, page 83
The Basics of MicroStrategy Objects, Object

Definitions, and Functions
To understand functions and their role in MicroStrategy, it is important to
grasp the basic concepts underlying objects and expressions.
An object is a basic building block in MicroStrategy. There are three types of

objects: schema objects, application objects, and configuration objects.
Schema objects include facts, hierarchies, and custom groups; application
objects include reports, documents, and metrics; and configuration objects
include project sources, database instances, and users. (This list of objects
is not exhaustive.)
MicroStrategy objects are created, maintained and deleted by you. Your

Architect, for instance, will create attributes by mapping conceptual data
from your data warehouse to names like Month, Customer Name, or Product
Category that will appear on your reports. Your Designer will create metrics
that access and manipulate numeric data from your data warehouse.
Many objects require you to specify an expression when creating or

modifying them. An expression is any combination of characters that can be
used as a result. Examples include the following:

l Month + 5
l Sum(Revenue)
l New York AND Los Angeles
In the first example, each Month element in your data warehouse is assigned
a numeric ID between 1 and 12, where January is 1, February is 2, and so
on. You want to generate a report with a five-month forecast. One approach
would begin with creating a new attribute called 5 Months Ahead, using the
expression Month + 5 in its attribute definition.
Expressions can be used in many places. Metrics in particular often require

complicated expressions. In the second example above, the expression Sum
(Revenue) is used to define a relatively simple metric. "Sum" tells
MicroStrategy to read individual Revenue entries from the data warehouse
and add them together to produce one single number that will be displayed
as Revenue on your report. This metric is a MicroStrategy object that you
can create.
The last example above shows a logical expression, which might be used in
a filter. A filter comprised of the expression New York AND Los Angeles
would let a user answer questions such as "Which business travelers flew
out of New York and out of Los Angeles in 2006?"
Expressions often form the basis of objects, which are the basic building
blocks of all MicroStrategy content and functionality. Objects that rely on
these expressions are often edited by changing the object expressions.
In this book, the terms "object expression" and "object definition" are
interchangeable unless otherwise noted.
To create an expression that accomplishes your goal, you will almost always
utilize a MicroStrategy function. Functions in MicroStrategy are powerful
tools used to define MicroStrategy objects (when they are integrated into
object definitions) and initiate complex user-selected calculations. The
example Sum(Profit) in a metric definition uses the Sum function to add

various Profit entries in a data warehouse to arrive at one final number to

display on a report.
Wherever you use an expression, you can use a function. From creating
basic objects to building complex reports and analyzing data, you can create
custom expressions using a large library of functions that come with and are
supported by MicroStrategy. Although functions are most commonly used in
metric expressions, they are also used to define attribute forms,
consolidation elements, custom groups, filters, facts, subtotals, and
transformations, all of which are MicroStrategy objects.
Functions Commonly Used to Create Specific Objects

The following table lists the function types described above, several
functions belonging to each type, and the MicroStrategy objects that can be
created using those functions.
Function MicroStrategy Objects That Can Use

Function Examples
Type These Functions
• Attribute form
• Arithmetic
operators • Consolidation
( –, +, /, *)
• Custom group
• Abs
Single-value • Fact
• Round
• Metric
• Ln
• Subtotal
• Cos
• Transformation
• Avg
• Metric
Group-value • Count
• Subtotal
• Sum
• Rank
OLAP Metric
• MovingMin

Function MicroStrategy Objects That Can Use

Function Examples
Type These Functions
• NTile
• And
Logical Filter
• Or
• Between
Comparison Filter
• ApplyComparison
• Attribute form
• ApplySimple • Consolidation
• ApplyAgg • Custom group
Apply • ApplyOLAP • Fact
• ApplyLogic • Metric
• ApplyComparison • Transformation
• Filter
Function Syntax and Formula Components

You can use functions in any situations where you build an expression.
Although an expression can have unique characteristics, there is a basic
syntax for applying functions in expressions. The use of each function is
described in detail in the rest of this book. This section covers the basics of
functions and their key components.
Each function is designated by a function name. Functions operate on an

argument that can be a fact, attribute, metric, or constant, and the function's
behavior is often further specified using one or more parameters. Function
parameters are used to fine-tune the behavior of functions. Arguments
provide the inputs to functions.

In this guide, MicroStrategy functions are grouped into function types. This
section provides information on function types, as well as their parameters
and arguments.
This guide does not include information on Data Mining functions. This
information can be found in the Data Mining Services chapter of the
Advanced Reporting Help.
Function Types
MicroStrategy functions are classified into the following types:
l Single-value functions (also known as Non-grouping or Scalar

functions): These functions operate on each individual component of an
input variable or argument, resulting in an output element for each
component. Examples of this category are simple mathematical operators
(+, -, *, /), Abs, Accrint, Ddb, Cos, Ln, Round, Truncate, ApplySimple, and
so on. For details, see Single-Value Functions, page 11.
l Group-value functions (also known as Grouping or Aggregate functions):

These functions take one or more lists of values as input and generate a
single output value for each list. Examples of this category are Avg,
AvgDev, Correlation, Count, HomoscedasticTTest, Intercept, Slope,
StDev, Sum, ApplyAgg, and so on. For details, see Group-Value
Functions, page 13.
l OLAP functions (also known as Relative functions): These functions take

multiple elements from a list and return a new list of elements. Each
element is related to and dependent on one or more other elements in the
list, and the relative positions of elements within the list determines how
computation is performed. Examples include Rank, RunningSum,

MovingAvg, NTile, ApplyOLAP, and so on. For details, see OLAP

(Relative) Functions, page 14.
l Comparison operators: These operators compare single values or lists of

values, or compare a list to a threshold value. Examples of this category
are Between, Like, Greater than(>), Less than (<), ApplyComparison, and
so on. For details, see Comparison Operators, page 16.
l Logical operators: These operators provide basic comparisons and

return TRUE or FALSE values based on the evaluation of the formula. This
type of operator includes And, Or, and Not. For details, see Logical
Operators, page 17.
l Apply functions: These functions provide access to functions and

syntactic constructs that are not standard in MicroStrategy but are offered
by various relational database management system (RDBMS) platforms.
Each of the functions in this category substitutes for one of the function
types mentioned above and can be used wherever that type is used. For
example, ApplySimple can be used wherever a single-value function is
used. For details, see Apply (Pass-Through) Functions, page 17.
In this guide, lists of attributes, metrics, values, parameters, etc. within a

function are separated by commas. Your separator may vary depending on
your number and date format locale. For example, the separator is a comma
(,) in English and a semicolon (;) in German.
Single-Value Functions
A single-value function operates on each individual component of one or
more arguments, resulting in an output component for each set of input
components. Subtraction, addition, division, and multiplication operators (–,
+, /, *) are common examples of single-value functions. Other examples
include Abs, Cos, Ln, Round, Sin, Trunc, ApplySimple, and so on.
Single-value functions can be used to create facts, metrics, attribute forms,

consolidation elements, and transformations.

For example, using a single-value function that operates on four arguments,

where each argument is composed of a five-component list, returns five
components. In other words, the number of output components is equal to
the number of input components in each argument. A simple example using
numbers follows:
Using the single-value "+" (addition) function, A+B=C
Where
l A=1 (an argument containing one component whose value is 1)
l B=2 (an argument containing a single component whose value is 2)
l C=3 (the returned value, containing a single output component, whose

value is 3)
In the example above, A, B, and C each contain one component. More

generally, given variables D and E used as arguments in the addition
function, where D and E each contain a five-component list, D+E results in
a five-component list. Single-value functions need not take a single value as
an argument or even a single argument. Rather, the basic requirement for a
function to be categorized as single-value is that the number of output
components equals the number of input components of the arguments.
The following two examples illustrate the use of single-value functions in the
creation of a transformed fact and a compound metric. Transforming a fact
and creating a compound metric are similar in that both use a single-value
function to turn one or more lists of values into another list of values. They
differ in that, for a transformation, the single-value function must be applied
before a group-value function, while in a compound metric the single-value
function is applied after the group-value function.
Exam ple 1: Transform ed Fact

Avg(Abs([Account Transactions]))

Suppose Account Transactions is a list of the following values: -300.5, -7.7,

900, -80, and 2.2. The single-value function, Absolute, is applied to the list.
The result set is the absolute value of each element in the list: 300.5, 7.7,
900, 80, 2.2. It is important to note that the single-value function returns five
elements of output for five elements of input. Once the single-value function
has been applied, the group-value function, Avg, is applied to produce an
average of those values, 258.08. For more information on the Abs and Avg
functions, see Abs (Absolute Value), page 416 and Avg (Average), page
117.
Exam ple 2: Com pound Metric

Avg(Revenue){Quarter} - Avg(Cost){Quarter}
In this example, the group-value function Avg is applied to both the Revenue
and Cost facts in your data warehouse. First, MicroStrategy uses the list of
values for the two input variables Revenue and Cost to generate, using the
Avg function twice, two new variables each containing a single value. The
two resulting variables are stored as intermediate results. Next, the single-
value function "-" (subtraction) is applied by MicroStrategy to subtract one
intermediate result from the other, resulting in a single value for the metric.
For more information on the Abs and Avg functions, see Abs (Absolute
Value), page 416 and Avg (Average), page 117.
In both of the previous examples, both single-value and group-value

functions were used. The next section addresses group-value functions in
more detail.
Group-Value Functions
A group-value function takes one or more lists of values as input and returns
a single output value for each list. The existence of a GROUP BY clause in a
SQL statement indicates that you are using a group-value function.
The most common group-value functions include Avg, Count, Max, Median,
Min, Stdev, Sum, Var, ApplyAgg, and so on. First, Last, IRR, and NPV

functions also belong to this category, but they have an additional sort by
feature (for more information, see Common Parameters, page 27). Sort by
specifies the order that the values returned by an expression will appear on
a report. (For more information on the SortBy parameter, see BreakBy and
SortBy Parameters, page 27.)
Group-value functions can be used to create simple metrics, nested metrics,

and compound metrics, as well as in the calculation of subtotals. The
following examples illustrate their use.
Exam ple 1: Average

Avg([Employee Age])
In this example, the group-value function Avg operates on the argument

Employee Age, which is a list of the following elements: 27, 35, 32, 47, 43,
40, 30. The function reduces the seven elements of the input value to a
single output value of 36. For more information on the Avg function, see Avg
(Average), page 117.
Exam ple 2: Median

Median([Employee Age])
The only difference between Example 2 and Example 1 above is the fact that
the group-value function, Median, is used, instead of Avg. Again, the
function reduces the seven elements of the input value to a single output
value of 35. For more information, see Median, page 134.
OLAP (Relative) Functions

Online Analytical Processing (OLAP) functions are also referred to as
Relative functions because each element in a list of values is related to and
dependent on one or more other elements in the list, and the positions of the
elements within the list determine how computation is performed.

An OLAP function takes multiple elements from a list and returns a new list
of elements. Unlike group-value functions, though, the number of elements
in the input list and the number of elements in the output list remains the
same. Unlike single-value functions, the computation depends upon the
conditions set by the BreakBy parameter that defines when the calculation
restarts and the SortBy parameter that defines how the list of values is
sorted (see Common Parameters, page 27).
OLAP functions include Rank, all the functions with Moving as the prefix of
the name (for example, MovingDifference, MovingMin, MovingStdev,
and so on), all the functions with Running as the prefix (for example,
RunningAvg, RunningCount, RunningSum, and so on), and all the NTile
functions (such as NTile, NTileSize, NTileValue, and
NTileValueSize). ApplyOLAP also belongs to the OLAP category.
OLAP functions are only used in the creation of metrics. The following is an
example.
Exam ple: RunningSum

RunningSum <BreakBy={[Customer Region]}, SortBy= ([Customer
State]) >(Revenue)
BreakBy refers to the attribute or hierarchy where calculations for an OLAP

function restart. To break by an attribute or hierarchy means to restart
calculations that use OLAP, or Relative, functions when the analytical
engine reaches the next instance of the specified attribute or hierarchy.
Examples of OLAP functions include RunningStdevP, Rank, NTile, and
various expressions that calculate percent values. To break by an attribute
or hierarchy in an expression, you set the BreakBy parameter.
The RunningSum metric computes the sum of the revenue for each
Customer Region by adding the revenue for each Customer State to the
revenue of the Customer States in the rows above it and displaying the
incremented total. The BreakBy Customer Region condition causes
calculations to begin again, however, when the next Customer Region is

encountered. (Notice in the figure below that the metrics Total Revenue and
the Running Sum for Arizona are equal because the calculation for Running
Sum has restarted, since Arizona is categorized in Southwest, a different
Customer Region than Wyoming, the previous Customer State on the
report.) Because of the SortBy Customer State condition, the Customer
States are listed in ascending (alphabetical) order, as shown in the report
excerpt below.
Comparison Operators
Comparison operators allow you to compare values. Using these operators,
you can compare single values or lists of values, or compare a list to a
threshold value.
Comparison operators include < (less than), > (greater than), = (is equal to),
Between, Contains, Ends with, ApplyComparison, and so on. They are only
used to create filters, which limit report data to a subset based on your need.
Exam ple: > (Greater Than)

Revenue > 500000

In this example, the filter limits the states in your yearly income report to
those with accrued revenue greater than $500,000.
Logical Operators
Logical operators provide basic comparisons and return TRUE or FALSE
values based on the evaluation of the formula. For numeric values, 0 is
treated as FALSE, and 1 is treated as TRUE. These operators provide a
means to combine data evaluations and comparison operators into complex
expressions. These expressions, in turn, can answer questions such as
"Which of our regions produced revenue that exceeded a 'success'
threshold?"
Logical operators include And, Or, Not, and ApplyLogic, all of which can only
be used to build filters where criteria are provided for the inclusion and
exclusion of data from a report display or metric calculation.
Exam ple: And

((Revenue - Cost) > 50000) And [Sell-through Percentage] > 25
Built for the attribute State, this filter limits report data to those states where
Profit (defined as Revenue - Cost) is greater than $50,000 and the Sell-
through Percentage is greater than 25%.
Apply (Pass-Through) Functions

The terms Apply functions and Pass-through functions are interchangeable.
They both denote functions in MicroStrategy that provide access to functions
or syntactic constructs that are not standard in MicroStrategy but are
provided by various Relational Database Management System (RDBMS)
platforms. The name "Pass-through" derives from the fact that MicroStrategy
passes information to a database which then uses its own functions. (Using
the native functionality of your RDBMS via Pass-through functions requires
that you know the syntax of your particular RDBMS. That syntax is beyond
the scope of this book and will vary from RDBMS to RDBMS.) RDBMS

functions, while necessary, must be used with care, since they always
bypass MicroStrategy's parsers and validators.
There are five predefined Apply functions that can be used to replace
regular or predefined functions of the same type. The functions are as
follows:
l ApplySimple: These functions are used where simple (for example,

arithmetic) operators can be used.
l ApplyAgg: These functions are used where aggregate functions (for

example, Sum) can be used.
l ApplyRelative: These functions are used where Online Analytical

Processing (OLAP) functions (for example, Rank) can be used.
l ApplyComparison: These functions are used where comparison operators

(for example, >, =, Like and In) can be used.
l ApplyLogic: These functions are used where logical operators (for

example, AND, OR, and NOT) can be used.
With Apply functions, project designers can customize expressions in the

Attribute, Filter and Metric editors to utilize RDBMS functions that are not
provided by MicroStrategy.
MicroStrategy strongly advises against using Apply functions when

standard MicroStrategy functions can be used to achieve the same goal,
because using RDBMS functions effectively bypasses the validations and
other benefits of MicroStrategy products. Using Apply functions is
recommended only when corresponding functionality does not exist in
MicroStrategy. When you need to use an Apply function, MicroStrategy
encourages you to submit an enhancement request for inclusion of the
desired feature in a future product release.

Exam ple: ApplyCom parison Used to Check a Prom pted Date
In this example, a table in your data warehouse contains the columns Item,
Effec Date, and Term Date (as well as Revenue), as shown below:
Item Effec Date Term Date Revenue
Blouse 06/01/2007 07/30/2007 1000
Jeans 05/30/2007 06/17/2007 500
Gloves 10/01/2007 10/25/2007 150
Leather
06/15/2007 06/22/2007 750
Shoes
Winter Hat 11/01/2007 11/08/2007 900
Winter
12/01/2007 12/15/2007 2200
Boots
Each row in the table corresponds to an item that was on sale during the
time between Effec Date and Term Date. Your objective is to generate a
report that lists all items (and an associated metric that you choose) that
were on sale on a particular date your user chooses at run time. To generate
this report, first create a value prompt named Test Date that allows the user
to input a date. Next, using that prompt, create a report filter using the
Custom expression box located in the Advanced Qualification pane of the
Filter Editor, as shown below.

Even though the filter is validated when you click Validate, MicroStrategy
returns an error when the report is executed. The error results from the fact
that you are supplying the SQL engine with two attributes and a value
prompt, while MicroStrategy is expecting to compare an attribute to the
attributes Effec Date and Term Date. In effect, you have a "type mismatch"
problem.
In this case, you can use an Apply function. Instead of having MicroStrategy
test the date value prompt, you instruct your database to perform the test. It
is important to remember that you have chosen to use an Apply function only
because MicroStrategy does not have a built-in function to accomplish your
task. If an appropriate MicroStrategy function existed, you would have used
it instead of an RDBMS function, because the latter does not offer the
validating features that MicroStrategy does. (To use Apply functions, you
must know the syntax of the corresponding function or operation in the
RDBMS you are using.)
To test the date prompt, use a custom expression to pass three values to the
database for comparison: the value prompt Test Date, the attribute Effec
Date, and the attribute Term Date. All three of these values are passed to
the database using placeholders in the form of #n, where n is a positive
integer that increases by 1 for each successive item being passed, starting
with 0. The first value passed is referred to as #0, the second is #1, the third

#2, and so on. The Custom expression in the Advanced Qualification pane of
the Filter Editor depicted below shows the syntax needed for this example:
Notice that the syntax is nested. The outer portion of the expression
contains the MicroStrategy function ApplyComparison, as well as the
MicroStrategy prompt Test Date and the attributes Effec Date and Term
Date.
The inner portion of the syntax, which is contained within double quotes, is
the database operation #0 between #1 and #2. Code that is passed to
the database using an Apply function is always enclosed in quotes, and the
arguments that are passed with that code are written as placeholders in the
form of #n, with the specific forms of the passed attributes specified by the
characters after the "@" sign. In this example, [Effec Date]@ID specifies
that MicroStrategy pass the ID form of the Effec Attribute instead of the
DESC or any other form that may exist in the database. At run time, #0, #1,
and #2 are replaced by Test Date, Effec Date, and Term Date, respectively,
so that the database effectively receives the following syntax:
Test Date between Effec Date and Term Date
If the user chooses 06/16/07 as the value of Test Date at run time, the
RDBMS reads the table row by row to see if the date falls between Effec
Date and Term Date. Whenever 06/16/07 falls between Effec Date and Term

Date on a particular row, the item on that row is returned in the result set. In
this example, Blouse, Jeans, and Leather Shoes are returned. (You can
verify this result by looking at the data warehouse table shown in the
beginning of this section.) If your report is set up with Item as a row
attribute, those three items appear on your report, indicating that they (and
only they) were on sale on 06/16/07.
For additional information about Apply functions, see Apply (Pass-Through)

Functions, page 171. The syntax of each Apply function as well as examples
appear in the sections that immediately follow.
Exam ple: Test Whether Hire Date is in the Current Year
Your HR department requires a list of employees that have been hired

during the current calendar year. The following custom expression uses the
ApplySimple function to test whether the year of Hire Date is the same as
the current year:
ApplySimple ( "datepart(yy, #0)", [Hire Date]@ID) =

ApplySimple ( "datepart(yy, getdate())", [Hire Date]@ID)
Each piece of the custom expression is explained below. More detailed

information on Apply functions in general can be found in Apply (Pass-
Through) Functions, page 17. More information on ApplySimple functions,
specifically, is found in ApplySimple, page 175.
l The datepart function extracts a specified part of a given date. The first
datepart function extracts the year (as directed by yy) from the ID
attribute form of the Hire Date attribute. The ID attribute form—as opposed
to the DESC or any other attribute form—is specified by @ID.
l The placeholder, #0, stands for the argument [Hire Date]@ID that is
passed to your RDBMS. (Apply functions use your database's
computational capabilities instead of those of MicroStrategy.)

l The second datepart function extracts the year (as instructed by yy)
from the current, or system, date. The system date is obtained via the
RDBMS function getdate().
l Your RDBMS extracts the year from both the Hire Date and the system
date, with MicroStrategy passing information to it. The container that
hands the necessary function to your RDBMS is an Apply function,
ApplySimple. In other words, ApplySimple acts as an interface between
you and the database, and when the RDBMS returns both year values,
they are compared with the = operator. If the year of a particular Hire Date
element is the same as the year of the system date, the custom expression
statement evaluates as true and that Hire Date attribute element is
returned in the result set of your report. If the year of a particular Hire Date
element is different than the year of the system date, the custom
expression statement evaluates as false, and that Hire Date attribute
element is filtered out of the report.
The attribute Hire Date is enclosed in brackets. Any time you type an
attribute whose name contains one or more spaces, the attribute must be
enclosed in brackets. (The use of brackets around compound object names
is standard for many objects in MicroStrategy and is not restricted to custom
expressions and Apply functions.)
The above example used an Apply function, ApplySimple. The next example
uses ApplyComparison.
Exam ple: Custom er City = Call Center Using ApplyCom parison
You need a list of customers who live in the same city as one of your call
centers. While it is possible to generate this report with a custom expression
that does not use an Apply function, this example uses an ApplyComparison
function to demonstrate Apply functionality within the custom expression.
(For steps to create this report without the use of an Apply function, see the
Attribute-to-attribute qualifications section of the Advanced Filters chapter
of the Advanced Reporting Guide.)

The custom expression used here evaluates whether one attribute is exactly
the same as another:
ApplyComparison ("#0 like #1",

[Customer City]@DESC, [Call Center]@DESC)
Each piece of the custom expression is explained below:
l The ApplyComparison function is used with RDBMS comparison

operators, such as the like operator used in this example.
l #0 like #1 is the actual comparison, comparing the first argument, #0,

with the second argument, #1. Remember that this comparison is done by
your RDBMS—not by MicroStrategy.
l [Customer City]@DESC sets the first argument passed to your RDBMS

as the description form of the Customer City attribute, while [Call
Center]@DESC sets the second argument passed to your RDBMS as the
description form of the Call Center attribute.
The attributes Customer City and Call Center are enclosed in

brackets. Any time you type an attribute whose name contains one or more
spaces, the attribute must be enclosed in square brackets. (The use of
brackets around compound object names is standard for many objects in
MicroStrategy and is not restricted to custom expressions and Apply
functions.)
Placing a filter that uses the custom expression above on a report that lists
the Customer City, Customer, and Call Center attributes yields the results
below. (Only a portion of the report is shown. Also, the Revenue metric has
been added.)

Notice that the custom expression qualification filter accomplished the goal
of returning only data that satisfies the criterion that the Customer City
attribute is the same as the Call Center attribute.
Function Parameters
Parameters determine how functions perform calculations. Any function,
whether standard or user-defined, can have parameters, which are
contained within angle brackets <> in an expression. If more than one
parameter is used, they are separated by commas.
Function parameter notations are only displayed in the Developer interface

if the parameter settings are changed from the default and you have set
your View option to Show Function Parameters.
You define parameters in the Function Name Parameters dialog box, which
displays the tabs Parameters, Break By, and Sort By. For steps on how to
access this dialog box and how to set function parameters with the Insert
Function Wizard, see Accessing and Modifying Function Parameters, page
29.

Since every function object has parameters, the Parameters tab is always
displayed. If a function has additional parameters, such as BreakBy and
SortBy, the related tabs are displayed accordingly. The following
subsections discuss the three tabs in more detail:
l Common Parameters, page 27
l BreakBy and SortBy Parameters, page 27
The details of each function are covered in this guide, including a listing and
description of each parameter that is available for the function. All
parameters are listed within angle brackets <> as part of the function syntax.
To review the details on the parameters available for each function, review
the functions provided in:
l Chapter 2, Standard Functions
l Chapter 3, Operators
l Chapter 4, Plug-In Package Functions

(,) in English and a semicolon (;) in Geman.

Common Parameters
There are three common Parameter settings for a function object:
l Distinct: is a TRUE/FALSE parameter that allows you to use all values or

only the unique values in the calculation.
l Fact ID: forces the calculation to take place on a fact table containing the
Fact_ID.
l NULL: is a TRUE/FALSE parameter that determines if the NULL value can

be used in the calculation.
Not all these settings apply to every function. In general, all group-value
functions display the FactID Parameter setting. OLAP functions do not have
any of these Parameter settings, but they may have BreakBy or SortBy
parameters.
BreakBy and SortBy Parameters

In addition to the Parameters settings, many functions have BreakBy or
SortBy parameters, each of which has its own individual settings:
l BreakBy: The logical level where the calculation of values for an

expression restarts. To break by an attribute or hierarchy means to restart
counting values for expressions that use relative functions. Examples of
relative functions are RunningStdevP, Rank, NTile, and expressions that
calculate rank or percent values. The break by level must at the same level
of aggregation or a higher level of aggregation used for the expression
itself.
For example, in the report shown below the Rank by Value metric ranks
the revenue values. The Rank function for this metric uses a BreakBy
of the Customer Region attribute, which means the rank calculation is
restarted for each customer region. This ensures that the revenue is
ranked by customer region, rather than ranking all of the revenues

together across customer regions. While ranking all the revenues

across customer regions can be useful, this report uses the BreakBy
parameter to focus on revenue comparisons within each customer
region.
l SortBy: The order of the return values of an expression in relation to the

order of the value or metadata object given. A sort by includes whether to
sort in ascending or descending order, and which metadata object to sort
by. Sort by may also be performed on the value of the subexpression,
which is the input argument.
For example, in the report shown below the FirstInRange metric returns
the first profit value in a list of profit values. The FirstInRange
function for this metric uses a sort by of the Customer State attribute,
which means the first value for each customer state is returned.

OLAP functions often include BreakBy and SortBy parameters. For example,
Rank has a BreakBy parameter, and MovingAvg has a SortBy parameter.
A few group-value functions (First, Last, IRR, and NPV) are also defined by
the SortBy parameter. The First and Last functions are used effectively to
calculate subtotals (see Subtotal Expressions, page 55).
For example, an inventory report lists the on-hand supply for each day. The
report subtotals are the last day's inventory. Creating a user-defined
subtotal that uses the Last function provides the last day inventory subtotal.
If the SortBy parameters of the function are not set to sort by Day, the
function may not provide the correct answer.
Accessing and Modifying Function Parameters

MicroStrategy Developer provides different methods for accessing and
modifying function parameters. You can access and modify function
parameters from the following interfaces:
l Insert Function Wizard: This interface is used to help guide you in the
initial inclusion of a function in an expression for a MicroStrategy object.
The Insert Function Wizard can be used to build a function and add it to an
expression for metrics, attributes, facts, subtotals, and transformations.

For more information, see Adding Functions to Expressions with the Insert
Function Wizard, page 57.
l Function Name Parameters dialog box: This interface is used to modify

the parameters of a function that has been added to an object expression
and validated.
To Access and Modify Function Parameters with the Insert Function

Wizard
1. Open an object editor for a MicroStrategy object that can include

functions in its expression.
For steps to access the Insert Function Wizard from the different
object editors, see the MicroStrategy online help and search for "Steps
to access the Insert Function Wizard". You can also see the section
Using Functions in Expressions, page 35 below that discusses the
different types of expressions in MicroStrategy.
2. Click the Insert Function button, labeled as f(x) on the expression

toolbar.
The Insert Function Wizard opens.
3. Use the Next and Back buttons to step through the pages of the Insert
Function Wizard. Each page allows you to modify different function

parameters.
The pages that you see depend on the function that you select.
To Access and Modify Function Parameters with the Function Name

Parameters Dialog Box
This functionality is available in any editor in Developer where functions are

used in expressions. For more information, see the following sections:
l Metric Expressions, page 36
l Attribute Form Expressions, page 45
l Custom Group Expressions, page 49
l Fact Expressions, page 51
l Filter Expressions, page 53
l Subtotal Expressions, page 55
l Transformation Expressions, page 56
1. Insert a function into an object expression and validate the expression.
You can insert a function by writing the name of the function and all
required parameters into the expression. You can also insert a function
using the Insert Function Wizard (see the procedure To Build an
Expression Using the Insert Function Wizard, page 57).
2. Highlight the function name, for example RunningSum, within a

validated expression in the expression box.
3. Right-click the function name and select function name parameters,

as demonstrated in the image below.

The Function Name Parameters dialog box opens. All parameter

tabs for the selected function are available, and settings editable,
from this dialog. This example uses the RunningSum function.
Examples of Function Parameter Effects

The following examples briefly illustrate the impact of parameters on
function execution.
Exam ple 1: Distinct Param eter

Count<Distinct=True>(Order)
In this metric, you modify the default parameter setting to Distinct=True and
retrieve a count of only the unique Orders.
Exam ple 2: RunningSum

RunningSum<BreakBy={Quarter}, SortBy=(Region)>(Revenue)
This metric is defined to display the running sum of revenue on a quarterly

basis, sorted by region in ascending order. For another RunningSum
example and its report, see OLAP (Relative) Functions, page 14.
Arguments
Arguments are the input data used in the calculation of a function.
Arguments can be numbers, text, or logical values (such as TRUE or FALSE)
as well as constants (such as 1, 2, 3, or NULL). They can be lists of values
or variables referencing lists of values.
Arguments in MicroStrategy are most often references to lists of values. In

function syntax, the arguments are enclosed in parentheses (). If the
argument is a reference to a MicroStrategy object, and the object name is
alphanumeric or contains multiple words, it is also contained in brackets [ ].
Depending on the function selected and the object being created, in a
MicroStrategy environment the input could comprise one or more of the
following objects:
l Attributes: Attributes are most often used to group fact data. They are
included in reports to define the level of detail. Typically non-numeric,
some common examples of attributes are Year, Category, and Region.
l Facts: Facts are the most commonly used input for metrics. They are
numerical lists obtained from specific columns in a fact table. Examples of
facts include Units Sold, Units Received, and Discount.
l Metrics: Metrics represent calculations performed on data and can

themselves be used as input for further calculation by a function.

Examples of metrics include Percent Growth, Profit Margin, and Sell-

through Percentage.
l Columns: Column data is used when creating attribute form expressions

and fact expressions. The expressions for these objects define how
column data is retrieved from the warehouse. Examples of columns
include TOT_DOLLAR_SALES, TOT_COST, and CUST_CITY_ID.

For an in-depth discussion of attributes, facts, and metrics, see the

Using Prompts for Arguments

Prompts can be used to provide the value for an argument in a function,
which allows a user to determine part of the function definition when a report
is executed. Arguments that expect a single value are the most common
arguments to use prompts on. For example, the function NTileSize (see
NTileSize, page 271) has the following syntax:
NTileSize <Ascending, BreakBy> (Argument, Size)
The two arguments for this syntax are:
l Argument is a metric representing a list of values to be distributed in

buckets.
l Size is a positive integer that designates the number of elements per

bucket.
The argument Size is a good candidate to use a prompt for, since it expects
a single value. Using a prompt to provide the value for Size allows a user to
determine how many elements should be included in each NTile bucket. This

is a more flexible reporting solution than defining a static value for the
argument that is always used for the calculation.
To include a prompt in a function expression, you can use the following

syntax:
?[Prompt Name]
The Prompt Name is the name of the prompt object. For example, you can
have the following definition:
NTileSize([Total Revenue], ?[NTileValue Prompt])
In the syntax shown above, NTileValue Prompt is the name of a value

prompt that supplies the size for the NTileSize function.
When prompts are created, you can choose whether answering the prompt
is optional or required. Since arguments are required for a function to work
properly, it is a good practice to define prompt answers as required if the
prompt is going to be used in a function expression.
Many of the financial functions (see Financial Functions, page 347) use
arguments that expect a single value, and thus are good candidates for
using functions to provide their values. For example, the function Coupdays
(see Coupdays (Coupon Period, Number of Days with Settlement), page
353) includes a Frequency argument which can accept the value of 1, 2, or
4 to determine the number of coupon payments per year. You could use a
value prompt for the Frequency argument to prompt the user to enter the
frequency of the coupon payments.
Using Functions in Expressions

Functions are the basis for many MicroStrategy objects. Some of the objects
they are used to create includes:
l Metric expressions
l Attribute form expressions

l Consolidation expressions
l Custom group expressions
l Fact expressions
l Filter expressions
l Subtotal expressions
l Transformation expressions
l Derived elements
l Derived attributes
This section explores the various roles of functions for each type of object
and how to access the expression builder in each case.
Metric Expressions
Metrics are MicroStrategy objects that represent calculations performed on
data. You can define metrics by using the available functions to analyze your
data and determine business measures.
Formula and dimensionality are the two important components in a metric.

While all metrics have a formula, not all of them have dimensionality.
l Formula: Is a mathematical expression using one or more functions,

applied to the data to be used in the calculation (facts, attributes,
constants, or metrics). In SQL, the formula commonly becomes part of the
SELECT clause of the SQL command.
You can re-use the same formula in multiple metric definitions. This type
of formula is called a base formula, which can contain arithmetic
operators, attributes, facts, group functions, and non-group functions. A
base formula does not have dimensionality (see below). For more
information on base formulas, see the Basic Reporting Help.
l Dimensionality: determines the attribute level at which a metric is

calculated. After deciding on the target (the attribute), in dimensionality

you can further define filtering and grouping involved in the metric. All
metrics, by default, calculate at the report level.
Other optional components of a metric include condition (filter) and

transformation. For the purposes of this book, we only discuss formula and
dimensionality related to the use of functions. For information on all metric
components, including additional information and examples of level
metrics, conditional metrics, and transformation metrics, see the
There are two types of metrics:
l Simple metric: has a formula and dimensionality (level). It can stand alone
or be used as a building block for a compound metric. A simple metric
must use at least one group-value function, such as Sum or Avg. It can
also contain a non-group function or arithmetic operator, in addition to the
required group function, for example, Sum(Revenue - Cost){~+};
however, the outermost formula must be a group function.
l Compound metric:isa combination of expressions that, through the use of

functions, are themselves metrics. In other words, a compound metric is
made of more than one complete metric. Any metric that is not a simple
metric is a compound metric by default. For example, all arithmetic
functions that are used as the root to connect two metrics yield compound
metrics.
A compound metric cannot have dimensionality placed on the entire

metric, although dimensionality can be set separately on each of its
component metrics.
A quick way to check whether a metric is simple or compound is through the

Metric Editor, where you can click the Subtotals/Aggregation tab and check
if the Allow Smart Metric option is enabled. If it is, then it is a compound
metric; if not, it is a simple metric.
The following three types of functions can be used to build simple and
compound metrics:

l Single-value functions
l Group-value functions
l OLAP functions
The optional condition (filter) component of a metric can contain logical and
comparison operators. See the Filter Expressions subsection for details.
While the single-value and group-value functions are used to create both
simple and compound metrics under different circumstances (see examples
to follow), the OLAP functions always yield compound metrics, due to their
unique characteristics (see OLAP (Relative) Functions, page 14).
Examples of Dimensionality in Metrics

As mentioned previously, all metrics have a formula, but not all metrics have
dimensionality. The following examples and diagrams illustrate the
formation of simple and compound metrics, as well as formula and
dimensionality.
Exam ple 1: Sim ple Metric with Dim ensionality at the Report Level
Avg(Revenue) {~+}
In this example, the expression is Avg on the fact Revenue. Together they
make up the base formula. This simple aggregation metric has
dimensionality, which is indicated by {~+}, meaning that the metric is

calculated at the lowest level on the report. For example, if a report contains
revenue by year and month, the numbers are calculated to reflect monthly
sales data. If month was removed from the report, the metric would
automatically be calculated at the new report level, which would reflect
yearly sales data.
All group-value functions are aggregation functions, which, when used

alone, yield simple metrics.
Exam ple 2: Sim ple Metric with Dim ensionality at Year Level
Sum(Cost) {~ +,year +}

In this example, the expression is Sum on the fact Cost. Together they make
up the base formula. This simple aggregation metric has dimensionality.
However, unlike in Example 1, dimensionality is set at the level of the
attribute Year, indicated by {Year +}. This means that if a report contains
cost by year and month, the numbers are calculated to reflect yearly cost
data.
Exam ple 3: Dim ensionality for Com pound Metrics

Avg(Revenue){~ +}/ Sum(Revenue) {~ +}

Dimensionality of compound metrics is defined by the other metric

definitions that are combined to create a compound metric. In this example,
the root Division (/) has two children, Avg(Revenue) and Sum(Revenue),
both of which are simple metrics themselves, and each of which has its own
dimensionality. The whole expression itself is a compound metric because it
uses two metrics and does not have its own dimensionality.
Accessing Metric Functions

You can access functions to create metrics in several ways. You can use the
Metric Editor when creating a new metric in a project. To create a derived
metric, which is a metric based on the existing data in a report, use the
options from within a report to insert a new metric.
You can also create metrics in Command Manager. For more information on
this method, see the Advanced Reporting Help.
Metric Editor
The Metric Editor is used to create new metrics and edit existing metrics in
MicroStrategy. The interface allows you to build metric expressions and

validate them.
To Access the Metric Editor Using Developer
1. Log in to a project.
2. In the MicroStrategy Developer File menu, point to New > Metric. The
New Metric dialog box is displayed.
3. Choose a Metric template and click OK. The Metric Editor displays.
4. Build the metric expression, accessing the functions in one of the

following ways:
l Expand the Functions and Operators folder using the drop-down list
or shortcut list in the Object Browser pane. Then expand the
Functions, Operators, or Plug-In Packages folder to access the
various categories of functions and operators.
l
Click (Insert Function) in the Definition pane. The Insert Function
wizard opens.
l Enter the function name and all required metric syntax directly in the
Enter your formula here box.
You can also edit existing metrics by using the Metric Editor. To
access the editor, select the appropriate metric in the folder list,
object browser, or report view. Then right-click and select Edit from
the shortcut menu. The Metric Editor opens with the selected metric
loaded.
To Access the Metric Editor Using Web
2. From the MicroStrategy Web home page, click Create Metric.

3. In the pane below, select the function to use to calculate data in the
metric. You can narrow the list of functions displayed in the pane by
doing one of the following:
l To search for the function by name, type the function's name in the
search field.
l Choose a function category from the drop-down list, such as Math

Functions or Financial Functions. The pane is updated to include only
the functions that belong to the selected category.
4. The Function Editor opens, with different options available depending

on the type of function you selected above:
l If you selected a grouping function, such as Sum, Average, First, or

Maximum, you are presented with options to define the metric's
expression, as well as optional components such as the level,
condition, and transformation. Perform the following steps:
1. Define the metric's expression by doing one of the following:
l To specify the expression by typing the name of an object,

type the name of the object in the Expression field. As you
type, matching objects are displayed in a drop-down list. You
can click an object or continue to type. You can type multiple
objects, such as Revenue-Profit.
l To specify the expression by choosing an object, click the

Browse icon. The Select an Object dialog box opens. Navigate
to and select an object, or search for the object.
2. You can further define the metric by adding a level, condition,

and transformation.

l If you selected a non-grouping function, such as data mining,

date, OLAP, and ranking functions, you are presented with
options to define the input values (called arguments) for the
function, as well as any parameters you can use to determine
the behavior of the function. For example, the NTile function
has two parameters, Ascending and Tiles. Ascending controls
whether the NTiles are ordered in ascending or descending
order, while Tiles sets the number of splits. To view a list of
the arguments and parameters for the function, click Details at
the bottom of the dialog box.
Perform the following steps:
1. For each argument listed, type a value or click the

Browse icon to find the metric, fact, prompt, or other
compatible object to use as input values of the function.
2. For each parameter listed, type a value or select the

parameter value from the drop-down
5. Click Save. For additional steps to define metrics using Web, see the
MicroStrategy Web Help .
Creating Derived Metrics
You create derived metrics based on objects already present in the report.
To Create Derived Metrics
1. Log in to the project.
2. Begin to create the derived metric:
l Using Developer, from the Insert menu in the Report Editor, select
New Metric. Or, right-click above a metric heading in the Report
Grid, point to Insert > New Metric. The Input Metric Formula dialog

box is displayed.
l Using MicroStrategy Web from the Data menu in the Report Editor,
select Insert New Metric. The Metric Editor is displayed.
3. Build a metric using the available report objects.
Attribute Form Expressions

Attribute forms are identifiers or descriptors of an attribute, such as ID,
Name, and Address. These units are part of an attribute, for example,
Customer. Attribute forms are defined by at least one expression, and these
expressions act on column data and can contain functions. The types of
attribute form expressions are as follows:
l Simple
l Implicit
l Derived
l Heterogeneous
For more information on these types of attribute forms, see the Advanced
Reporting Help.
In the context of MicroStrategy functions, this book discusses derived

expressions. A derived expression can only use single-value functions, and
arguments that are used in the expression are columns. See the examples
described below.
Example 1: Subtraction ( - )
(Year(CurrentDate()) - Year([HIRE_DATE]))
The attribute form Employee Experience is defined by the above expression

using the simple mathematical operator, subtraction.

This example can be found in the MicroStrategy Tutorial project in the

following folder: MicroStrategy Tutorial/Schema
Objects/Attributes/Geography.
Example 2: InitCap
InitCap([CUST_LAST_NAME])
Text data is typically in all upper-case or all lower-case letters. This example
shows that for the attribute form of customer's last name, you can use the
single-value function, InitCap, to make the first letter capitalized and all
other letters in lower case.
Example 3: ApplySimple
ApplySimple("Datediff(YY,#0,getdate())", [BIRTH_DATE])
The attribute form, Age, can be defined by using the single-value function,
ApplySimple.
l For all Apply functions, do not use a group for the attribute form
expression. Use a single form because form groups are ignored by the
Analytical Engine. For example, you cannot use Customer@Name, where
Name is a form group defined as the customer's first name, middle name,
and last name.
l The syntax of apply functions is database-specific. For more information,

see Internal Functions, page 169.
To Access Attribute Form Expressions
Fo r a N ew At t r i b u t e
1. From the MicroStrategy Developer File menu, point to New > Attribute.
The Attribute Editor opens three dialog boxes, if the cascading dialog

box option is enabled in the Developer Preference (as it is by default).

The three dialog boxes are New Attribute, Create New Attribute Form,
and Create New Attribute Form Expression. The Create New Attribute
Form Expression dialog box is where functions and operators are used.
2. Build your expression by using the functions in one of the following

ways:
l
wizard opens.
l Enter the function name and all required expression syntax directly in
the Enter your formula here box.
Fo r an Exi st i n g At t r i b u t e
1. To access the editor, select the attribute in the folder list, object
browser, or report view and right-click and select Edit.
2. Select an attribute form and click Modify. The Modify Attribute Form
dialog box displays.
To add a new form to the existing Attribute, click New. This

automatically opens the New Attribute Form Expression dialog box.
3. Select the expression you want to edit and click Modify.
To add a new expression to the existing form, click New. This

automatically opens the New Attribute Form Expression dialog box.
4. The Modify Attribute Form Expression dialog displays with the selected
expression loaded.
5. Create or edit the expression, accessing the functions in one of the

following ways:

l
wizard opens.
l Enter/edit the function name and all required expression syntax in the
Consolidation Element Expressions

Consolidations enable you to group attribute elements for use in a report,
without changing the structure of your metadata or your warehouse
definition. The elements contained in a consolidation are called
consolidation elements (CEs). Only single-value functions can be used in
the definition of a consolidation element as well as calculations between
consolidation elements. A consolidation element expression defines how the
attribute elements are calculated.
Only the basic mathematical operators (+, -, *, /) can be used through

MicroStrategy Developer to define consolidation elements. Examples of
consolidations, using these operators, are as follows:
l CE01 = {Region = North-East}
l CE02 = {Region = Mid-Atlantic}
l CE03 = ({CE01} – {CE02})/{CE02}
Other single-value functions can also be used for consolidation elements,

but only through the SDK.
The following examples demonstrate definitions using functions and

calculations:
l CE04 = Ln({Region = North-East})
l CE05 = Ln({Region = Mid-Atlantic})
l CE06 = Abs({CE04} – {CE05})

For more information on consolidations, see the MicroStrategy Advanced

Reporting Guide.
To Access the Consolidation Element Expressions
1. In the MicroStrategy Developer File menu, point to New >

Consolidation. The Consolidation Editor opens.
2. Double-click Click here to add new consolidation element.
3. The New Consolidation Element pane is enabled. Build the expression

for the new element.
You must drag and drop attributes into the Enter your expression here
box. Only the operators can be typed directly in the box.
Custom Group Expressions

A custom group is an object that can be placed on a template and is made up
of a collection of elements called custom group elements. A custom group
can group attribute elements in a way that is not defined in the data
warehouse. You can create relationships between the attribute and the
custom group. A custom group expression defines how the elements in the
custom group are calculated.
A custom group can organize attribute elements through:
l Attribute qualification
l Set qualification
l A report
l A filter
l Banding
l Advanced qualification

For more information on custom groups, see Custom Groups and

Consolidations in the MicroStrategy Advanced Reporting Guide.
When you define custom group elements through advanced qualification,

you can use two types of functions:
l Logical functions
l Comparison functions
See the following examples.
([Unit_Profit] - [Unit_Cost]) > 500
Example 2: And
([Units Sold] > 500) and ([Unit Profit] > 1000)
To Access the Custom Group Element Expressions
1. In the MicroStrategy Developer File menu, point to New > Custom

Group. The Custom Group Editor opens.
2. Double-click Double-click here or drag an object from the object

browser to add a custom group element.
3. Enter a name for the new element and double-click Add Qualification.
4. In the Custom Group Options pane, select Add an Advanced

qualification and click OK.
5. The Advanced Qualification pane opens. This is where the expression

is built. The functions are accessed in one of the following ways:

l
wizard opens.
l Enter the function name and all required expression syntax directly in the
Fact Expressions
Facts are objects created by and shared between MicroStrategy users. They
relate numeric data values from the data warehouse to the MicroStrategy
reporting environment. A fact expression defines how a fact is calculated.
The fact expression is part of the Fact Definition component of a fact
structure. You can use functions to create fact expressions, and arguments
that are used in the expressions are columns.
Facts can be defined as:
l Implicit
l Derived
l Heterogeneous
For more information on facts and fact structure, see the MicroStrategy
Project Design Guide.
In the context of MicroStrategy functions, this book discusses derived fact

expressions.
Derived fact expressions can only use single-value functions, including

simple arithmetic operators (+, -, /, *). See the following two examples.
([UNIT_PRICE] - [UNIT_COST])
The fact Unit Profit is defined using the table columns Unit_Price and Unit_
Cost and a simple mathematical operator, subtraction.

Example 2: Multiplication ( * )
([QTY_SOLD] * ([UNIT_PRICE] - DISCOUNT))
The fact Revenue is defined using three table columns and two arithmetic
operators, subtraction and multiplication.
Both examples can be found in the MicroStrategy Tutorial project in the

following folder: MicroStrategy Tutorial/Schema Objects/Facts.
To Access Fact Expressions
Fo r a N ew Fact w i t h an Exp r essi o n
1. In the MicroStrategy Developer File menu, point to New > Fact.
2. The Expression page is where the expression is built. The functions are
accessed in one of the following ways:
l
wizard opens.
l Type the function name and all required expression syntax directly in the
Fo r an Exi st i n g Fact
1. Select an existing fact in the MicroStrategy Developer main screen.

Then right-click and select Edit. The Fact Editor displays with the
selected fact loaded.
2. Click New to add a new expression to the fact, or select an existing

expression and click Modify. The Create New Fact Expression or
Modify Fact Expression page displays.
3. Build an expression by using functions in one of the following ways:

l
wizard opens.
l Enter the function name and all required expression syntax directly in the
You can use either the Fact Editor or the Fact Creation Wizard to create
facts; however, only the Fact Editor allows you to use advanced
expressions to define the fact.
Filter Expressions
A filter specifies the conditions that data must meet to be included in the
report results. In SQL, a filter is specified after the WHERE clause. A filter
can be a report object, that is, a report filter, a report limit, or a view filter,
that restricts the data returned or the display or view of data on a report. For
detailed information on report filters, see the Advanced Reporting Help .
A filter can also be a metric qualifier created using custom expressions

employing functions. It is used as the optional condition component of a
metric. You can create this type of filters by using the Advanced
Qualification dialog box within the Filter Editor. Only the following types of
functions can be used in filter expressions:
l Logical operators
l Comparison operators
Single-value functions can be used at the sublevel in a filter expression, as

long as the root node is a logical or comparison function (see Example 1).
Example 1: Greater Than ( > )

((Revenue - Cost) > "5000")
This example uses a simple comparison operator to create a filter to limit the
display of profit to values greater than $5,000.

You can achieve the same result by using a set (metric) qualification on a
compound metric (such as the one in the example defined as Revenue -
Cost). The custom filter expression is used here for illustration purposes. It
is up to you which method you prefer to use.
Example 2: ApplyComparison
ApplyComparison("#0 between #1 and #2", ?[Value Prompt Date],
[Order Date]@ID, [Ship Date]@ID)
ApplyComparison is commonly used to create custom filters. In this

example, the filter compares a user-entered date to see if it is between the
Order_Date and the Ship_Date.
To Access an Advanced Filter Qualification
1. In the MicroStrategy Developer File menu, point to New > Filter. The
Filter Editor is displayed.
2. In the Filter Definition pane, double-click Double-click here to add a

qualification or drag an object from the Object Browser. The
Filtering Options pane is displayed.
3. In the Filtering Options pane, select the Add an Advanced

qualification option and click OK. The Advanced Qualification pane is
displayed.
4. Select Custom Expression from the Option list. Then use the Custom
Expression box to build and validate your custom filter. Access the
functions in one of the following ways:
or folder list. Then expand the Functions, Operators, or Plug-In
Packages folder to access the various categories of functions and
operators.

l
Wizard opens.
l Enter the function name and all required metric syntax directly in the
Subtotal Expressions
Subtotals allow you to dynamically control the computation and display of
report data within desired groupings. Subtotals are applied to report metrics
to calculate totals and for dynamic aggregation.
The standard predefined subtotal functions, which are automatically

available for use with every metric and report, are simple aggregate
functions that satisfy many subtotaling requirements. If they do not answer
your particular needs, you can create a user-defined subtotalusing the
Subtotal Editor or through the SDK. User-defined subtotals allow you to
develop your own subtotals, using single-value functions or group-value
functions.
Both predefined and user-defined subtotals can be applied to reports from

the Subtotals/Aggregation tab in the Metric Editor.
For more information on subtotals and the procedures for creating and
applying them, see the Reports chapter of the Advanced Reporting Help or
the MicroStrategy online help.
Example: Division ( / )
Sum(Sum(x*[Units Sold]){Year}/Sum([Units Sold]){Year}){}
To Access Functions in the Subtotal Editor
2. In the MicroStrategy Developer File menu, point to New and then select
Subtotal. The Subtotal Editor displays.

3. The Subtotal Editor is very similar to the Metric Editor. To access the
functions used to create a new subtotal, you can do one of the
following:
or shortcut list in the Object Browser pane. Then expand the
Functions, Operators or Plug-In Packages folder to access the
various categories of functions and operators.
l
wizard opens.
l Enter the function name and all required subtotal syntax directly in
Transformation Expressions
Transformations are schema objects that use business rules to compare
values at different time periods. A typical example of this type of analysis is
a TY/LY comparison (This Year versus Last Year).
Transformations are schema objects; therefore, you must have the

appropriate privileges to create or modify them.
There are two types of transformations: expression-based and table-based.

Single-value functions can be used in expression-based transformations,
and specifically in the definition of member expressions. These expressions
define how (and from where) the information is retrieved for the
transformation of the specified attribute.
For more information on transformations and their components, see the

Advanced Reporting Help or the MicroStrategy online help.

To Access Functions in Transformations
1. In the MicroStrategy Developer File menu, point to New >

Transformation. The Transformation Editor is displayed. The Select
Member Attribute dialog box also opens.
2. Select the attribute on which to base the transformation and click OK.
The Expression Editor opens.
3. Build an expression for the transformation of the selected member

attribute, accessing functions in one of the following ways:
l
wizard opens.
l Enter the function name and all required expression syntax directly in
Adding Functions to Expressions with the Insert Function

Wizard
The Insert Function Wizard is available wherever functions are used. It
guides you through the process of selecting a function, choosing the data on
which the function acts, and setting the available parameters.
To Build an Expression Using the Insert Function Wizard
For detailed information on each page of the Insert Function Wizard, see the
MicroStrategy online help and search for "Using the Insert Function
Wizard."
1. Click the Insert Function button to access the wizard. The Select
Function page of the Insert Function Wizard is displayed.

2. Select a function and click Next. The Arguments page opens.
3. Enter the values or names of the arguments you want to add or browse
for the arguments:
l Enter the appropriate values or name of the arguments into the

available boxes.
l Click (the Browse button). The Open dialog box opens, which is a
standard Windows dialog box for file selection.
Select the argument on which the function acts. Click Open to select
the argument and return to the Insert Function Wizard.
4. Click Next. The Parameters page is displayed (if applicable).

For group-value metrics such as Count, Min, and Max, standard

parameters (Distinct, NULL, and FactID) display together on the only
page available, called Parameters. The default for FactID is (Nothing),
meaning that the calculation searches for the input argument from the
lookup table. Otherwise, make a selection from the pull-down list to
force the calculation to look in another table.
5. Set the values of the parameters.
6. Define the SortBy and BreakBy parameters in their separate windows, if

applicable.
7. Click Finish. The expression built through the Insert Function Wizard
displays in the appropriate editor.

Parameter values are displayed only if you have set your view to Show
Function Parameters and have modified a parameter from its default
setting.
How MicroStrategy Processes Functions

MicroStrategy Intelligence Server has an engine component, which
comprises the following:
l SQL Engine: Generates the SQL and associated logic for functions
performed by the database, and communicates with the Analytical Engine
as necessary.
l Query Engine: Sends the SQL generated by the SQL Engine to the data
warehouse for execution.
l Analytical Engine: Extends the capability of the system beyond what the
RDBMS provides. For example, it performs complex calculations on a
result set returned from the data warehouse, such as statistical and
financial functions, subtotal calculations on the result set, metric
calculations that are not or cannot be performed using SQL, such as
complex functions, and so on.
While this extends the capabilities of the RDBMS that you are using,
calculations that require the Analytical Engine can require additional
system resources and processing time. To determine if a function is
supported by your RDBMS or will be evaluated by the Analytical Engine,
review the function support listed in Appendix A, MicroStrategy and
Database Support for Functions.
Types of Function Processing

Functions supported by the Intelligence Server are of three types:
l Those that can be processed only by the Analytical Engine, such as

finance functions.

If the Analytical Engine does not support a given function, a compound

metric containing the function cannot be smart metric enabled. This is
because smart metrics change the default order of metric evaluation and
only the Analytical Engine can support such a change.
l Those that can be processed either by the Analytical Engine or by the

database.
If the database can perform the calculation, the SQL Engine sends the
instructions to the database; otherwise, the Analytical Engine processes
the task.
How Intelligence Server Uses Functions

Recall that there are several categories of functions recognized by
Intelligence Server. These function types include group-value, single-value,
and OLAP functions. Within these categories are functions supported by
only the Analytical Engine, only the database, or by both the engine and the
database. The SQL generated to process the request differs depending on
what processes the request. The SQL syntax also differs depending on the
database used, when the database supports the function.
The following subsections explore through examples of how the various uses
of functions result in different SQL syntax. All of these examples use the
functions in the context of metrics.
Most of these examples (except where noted) can be recreated using the
objects in the MicroStrategy Tutorial. Tutorial data is stored in a Microsoft®
Access database.
Using Group-Value Functions

The computation of group-value functions is done by either the Intelligence
Server or the database depending upon the function used and support
available. The following examples discuss how MicroStrategyperforms the
group-value computations by providing the SQL syntax for specific
situations.

Group-Value Functions in Simple Metrics

This subsection contains two examples that illustrate the processing of
group-value functions. Each example contains two reports, the first one
showing the SQL for a function supported by the database and the second
one by the Analytical Engine.
Exam ple 1: Sum (Revenue) vs. AvgDev(Revenue)
Consider a simple report, Report 1A, where the attribute Region is placed on
the row axis and a simple metric defined as M1A=Sum(Revenue){~} is on
the column axis. This report shows the sales for each region. The database,
in this case Microsoft Access, supports the function Sum. The following SQL
is generated:
l Report 1A (SQL Group-value function) - Microsoft Access
select a12.[REGION_ID] AS REGION_ID,

max(a13.[REGION_NAME]) AS REGION_NAME,
sum(a11.[TOT_DOLLAR_SALES]) AS WJXBFS1
from [CITY_CTR_SLS] a11,
[LU_CALL_CTR] a12,
[LU_REGION] a13
where a11.[CALL_CTR_ID] = a12.[CALL_CTR_ID] and
a12.[REGION_ID] = a13.[REGION_ID]
group by a12.[REGION_ID]
Now, consider Report 1B, again with the same attribute, Region, on the row
axis, but with a different simple metric defined as M1B=AvgDev(Revenue)
{~} on the column axis. This report shows how revenue data varies from its
mean for each region. The database does not support the function Average
Deviation; therefore, the computation is performed by the Analytical Engine.
The following SQL is generated:

l Report 1B (MicroStrategy Group-value function)
select a11.[CALL_CTR_ID] AS CALL_CTR_ID,

a11.[CUST_CITY_ID] AS CUST_CITY_ID,
a12.[REGION_ID] AS REGION_ID,
a13.[REGION_NAME] AS REGION_NAME,
a11.[TOT_DOLLAR_SALES] AS WJXBFS1
[LU_CALL_CTR] a12,
[LU_REGION] a13
[An analytical SQL]
In Report 1B, Intelligence Server performs the following steps:
l It retrieves all fact data from the warehouse. [TOTAL_ DOLLAR_SALES] is

the column alias used for the fact (Revenue) in the temporary table during
SQL generation.
l It uses the result set (that is held in memory) to compute the metric,
namely AvgDev([TOTAL_ DOLLAR_SALES]){~} for each region.
l The notation [An analytical SQL] indicates that the computation is

taking place in the Analytical Engine.
l It displays the final result.
In the previous examples, the dimensionality of both metrics is defined as

{~}, which means that they both are calculated at the report level of Region,
since the attribute Region is on the reports.
The next example explains how Intelligence Server processes

dimensionality. When metric dimensionality is defined, the Analytical Engine
can insert records back into the temporary database structures after the
function calculation is performed so that dimensionality can be applied.

Exam ple 2: Sum (Revenue) {~, Country} vs. AvgDev(Revenue) {~, Country}
Now use the same report template as in Example 1, but add a dimensionality
to each metric for comparison. Use Sum(Revenue){~,Country} for
Report 2A and AvgDev(Revenue){~,Country} for Report 2B. Notice that
attribute Country is a parent of Region, and the relationship is one to many.
In Report 2A, the Sum function is supported by the database; in Report 2B,
the AvgDev function is supported by Intelligence Server. The SQL generated
for both reports is as follows:
Report 2A (SQL Group-Value Function) - Microsoft Access
create table ZZT1Y03009ZMD000 (

COUNTRY_ID BYTE,
WJXBFS1 DOUBLE)
insert into ZZT1Y03009ZMD000
select a12.[COUNTRY_ID] AS COUNTRY_ID,
[LU_CALL_CTR] a12
where a11.[CALL_CTR_ID] = a12.[CALL_CTR_ID]
group by a12.[COUNTRY_ID]
pa1.[WJXBFS1] AS WJXBFS1
from [ZZT1Y03009ZMD000] pa1,
[LU_REGION] a11
where pa1.[COUNTRY_ID] = a11.[COUNTRY_ID]
drop table ZZT1Y03009ZMD000
The first pass of SQL creates a temporary table to hold the data. The second
pass computes the metric at the Country level, while the third pass joins with
attribute Region since the result of the aggregation has to be displayed for
each region. The final pass drops the temporary table.

Report 2B (MicroStrategy Group-Value Function)
select a11.[CUST_CITY_ID] AS CUST_CITY_ID,

a12.[COUNTRY_ID] AS COUNTRY_ID,
a11.[CALL_CTR_ID] AS CALL_CTR_ID,
[LU_CALL_CTR] a12
where a11.[CALL_CTR_ID] = a12.[CALL_CTR_ID]
create table ZZMD00 (
COUNTRY_ID BYTE,
WJXBFS1 FLOAT)
[An analytical SQL]
insert into ZZMD00 values ([DummyInsertValue])
from [ZZMD00] pa1,
[LU_REGION] a11
where pa1.[COUNTRY_ID] = a11.[COUNTRY_ID]
drop table ZZMD00
In Report 2B, an Analytical SQL pass is necessary to compute AvgDev since

it is not a database supported group-value function. In the next pass, the
results of the calculation are inserted back into the temporary database
structures. The last SQL pass is the same as Report 2A, since it is used to
display the result for all regions.
Using Single-Value Functions

The key to understanding the computation of a single-value function is to
identify the way it is used. The next two subsections provide examples of the
two uses of single-value functions and how they are processed.
The first example shows a single-value function applied before a group-

value function. This is referred to as transforming a fact. The second
example shows a single-value function applied after the group-value
function. This is referred to as a compound metric.

Each example contains two reports, the first one showing the SQL syntax
when calculations are processed by the database and the second one when
calculations are processed by the Intelligence Server.
The use of transform in this context refers to retrieving a fact in a different

form. For example, you can obtain the absolute value for a fact or the
natural logarithm, but the fact is the same. This differs from a
transformation where the data retrieved is different when a transformation is
applied, for example, last year's revenue instead of this year's revenue.
Transforming a Fact Into Another Fact

In this case, the fact Revenue is transformed into another fact, namely the
natural logarithm values defined as Ln(Revenue) or the truncated revenue
values defined as Trunc(Revenue).
The following example describes this type of usage and displays the SQL
that is generated.
Consider the two metrics defined as follows:
l M3A = Sum(Ln(Revenue)){~+}
l M3B = Sum(Trunc(Revenue)){~+}
Put M3A with Region on the template. The database, in this case, SQL
Server 2000, supports the function Ln. The following SQL is generated:
Report 3A (SQL Single-Value Before Group-Value Function) - SQL Server

2000
Pass0 - Duration: 0:00:40.45

select a13.REGION_ID REGION_ID,
max(a14.REGION_NAME) REGION_NAME,
sum(LOG(a11.ORDER_AMT)) WJXBFS1
from ORDER_FACT a11
join LU_EMPLOYEE a12
on (a11.EMP_ID = a12.EMP_ID)

join LU_CALL_CTR a13

on (a12.CALL_CTR_ID = a13.CALL_CTR_ID)
join LU_REGION a14
on (a13.REGION_ID = a14.REGION_ID)
group by a13.REGION_ID
Based on the SQL, this is what happened:
1. The Ln function is applied to the fact Revenue, which is defined in the

warehouse by the column ORDER_AMT.
The Ln function uses the syntax LOG in SQL Server 2000.
2. Then, the Sum function is performed on the new fact, namely Ln

(Revenue).
Now, put M3B with Region on the template, the following SQL is generated:
Report 3B (MicroStrategy Single-Value Before Group-Value Function)
select a11.[CALL_CTR_ID] AS CALL_CTR_ID,

a12.[REGION_ID] AS REGION_ID,
[LU_CALL_CTR] a12,
[LU_REGION] a13
[An Analytical SQL]
In this case, as noted by the text [An Analytical SQL], the functions
Trunc and Sum are computed by the Intelligence Server. Intelligence Server
computes the new fact Trunc([Dollar Sales]) first and then uses the
Sum function to sum the new fact for each region.

Calculating a Compound Metric

Single-value functions can be used to create compound metrics.
Subtraction, addition, division, and multiplication operators (–, +, /, *) are
common examples of single-value functions. See the following examples:
l (Sum([Dollar Sales]){~}/Sum([Dollar Sales]) {~,Country})
l (Sum([Dollar Sales]){~} + Sum([Freight]) {~})
The examples below use a compound metric and a metric that transforms a
fact in the same report. The examples illustrate the SQL generated when the
function is supported by the database and the Intelligence Server,
respectively.
Consider the following metric definitions:
l M4A = Ln(Sum(Revenue){~})
l M4B = Trunc(Sum(Revenue){~})
Put metric M3A, from the previous example, and metric M4A together with
attribute Region on the template. The database, in this case SQL Server
2000, supports the Ln function. The following SQL is generated:
Report 4A (SQL Single-Value Before and After Group-Value Functions) -

SQL Server 2000

select a13.REGION_ID REGION_ID,
max(a14.REGION_NAME) REGION_NAME,
sum(LOG(a11.ORDER_AMT)) WJXBFS1,
LOG(sum(a11.ORDER_AMT)) WJXBFS2
from ORDER_FACT a11
join LU_REGION a14

group by a13.REGION_ID
To process metric M3A, the single-value function Ln is calculated before the

group-value function Sum, and for metric M4A, Ln is calculated after Sum.
The Ln function uses the syntax LOG in SQL Server 2000.
Similarly, if you put metrics M3B and M4B together with attribute Region on
the template, and the database does not support the Trunc function, then the
following SQL is generated:

select a11.ORDER_DATE DAY_DATE,
a11.EMP_ID EMP_ID,
a11.ORDER_ID ORDER_ID,
a13.REGION_ID REGION_ID,
a14.REGION_NAME REGION_NAME,
a11.ORDER_AMT WJXBFS1
from ORDER_FACT a11
join LU_REGION a14
[An Analytical SQL]
The fact, Revenue, which is defined as column ORDER_AMT in the ORDER_

FACT table, is retrieved and used to compute both metrics.
Using OLAP Functions

You can better understand how Intelligence Server computes OLAP
functions by observing several examples based on the following properties:
l Window size
l BreakBy

l SortBy
l NULL handling
l Tie handling
There are two examples for each subsection, comparing the SQL syntax
when the database performs the OLAP function calculations with the one
when the Intelligence Server performs the computation.
Window Size in Moving Functions

For this example, define metric M1A as Sum(Revenue){~+}. Then create
an OLAP metric called OM1, that is defined as
MovingAvg <BreakBy = {Category}, SortBy = (Value)>([M1A],5).
OM1computes the moving average of M1A with a window size of 5. The

OLAP metric is computed after it is sorted by the value of M1A in ascending
order. Moreover, the computation restarts for every category.
Now, put the attributes Category and Item on the Row axis and the above
metrics on the Column axis. Run this report with an empty filter against a
database that does not support OLAP functions (in this case Microsoft
Access), the following SQL is generated:
Report 5A (MicroStrategy OLAP function - window size) - Microsoft Access

select a13.[CATEGORY_ID] AS CATEGORY_ID,
max(a14.[CATEGORY_DESC]) AS CATEGORY_DESC,
a11.[ITEM_ID] AS ITEM_ID,
max(a12.[ITEM_NAME]) AS ITEM_NAME,
from [ITEM_MNTH_SLS] a11,
[LU_ITEM] a12,
[LU_SUBCATEG] a13,
[LU_CATEGORY] a14
where a11.[ITEM_ID] = a12.[ITEM_ID] and

a12.[SUBCAT_ID] = a13.[SUBCAT_ID] and

a13.[CATEGORY_ID] = a14.[CATEGORY_ID]
group by a13.[CATEGORY_ID],
a11.[ITEM_ID]
[An Analytical SQL]
Alternatively, if you execute the above report against a database that does
support OLAP computations (in this case Oracle 9i), then the following SQL
is generated:
Report 5B (DB OLAP Function - Window Size) - Oracle 9i

select a13.CATEGORY_ID CATEGORY_ID,
a14.CATEGORY_DESC CATEGORY_DESC,
a11.ITEM_ID ITEM_ID,
a12.ITEM_NAME ITEM_NAME,
sum((a11.QTY_SOLD * (a11.UNIT_PRICE - a11.DISCOUNT)))
WJXBFS1,
avg(sum((a11.QTY_SOLD * (a11.UNIT_PRICE -
a11.DISCOUNT)))) over(partition by a13.CATEGORY_ID
order by sum((a11.QTY_SOLD * (a11.UNIT_PRICE -
a11.DISCOUNT))) asc rows 4 preceding) WJXBFS2
from ORDER_DETAIL a11,
LU_ITEM a12,
LU_SUBCATEG a13,
LU_CATEGORY a14
where a11.ITEM_ID = a12.ITEM_ID and
a12.SUBCAT_ID = a13.SUBCAT_ID and
a13.CATEGORY_ID = a14.CATEGORY_ID
group by a13.CATEGORY_ID,
a14.CATEGORY_DESC,
a11.ITEM_ID,
a12.ITEM_NAME
The following example shows how the Intelligence Server computes OLAP
functions when the database does not support OLAP functions. The

Intelligence Server retrieves all components: the input metric, BreakBy

parameter, and SortBy parameter. In the above examples, since the SortBy
parameter setting is by Value, it is sufficient to bring back just the input
metric ( WJXBFS1 ).
BreakBy in OLAP Functions

In the Intelligence Server, the BreakBy parameter is only available at the
attribute level. In other words, you can start over certain computations of
OLAP functions when a part of the metric belongs to a different attribute
element.
In the previous example (window size), there is an OLAP function with

Category in the BreakBy parameter. If the Analytical Engine computes the
OLAP function, Intelligence Server must select this attribute in the select
clause. If the database computes the OLAP function, then this attribute must
also be in the partition by clause.
Attributes in the BreakBy parameter of an OLAP metric are always applied,

meaning that they are always in the Select clause, and if the database
computes it, they are also in the Partition by clause. This is true
whether the attribute is on the template or not.
Run a report similar to the one for window size, deleting the attributes
Category and Item from the template and adding Subcategory. The results
for metric M1A are completely different from the previous report because if
Category and Item are not on the template, then the level of aggregation for
metric M1A is replaced by Subcategory, which is on the template. On the
other hand, OLAP metric OM1 must still restart the calculation (break by) for
each Category and therefore remains in the appropriate Select and
Partition by clauses. Notice that in both the SQLs below, the
Intelligence Server always selects Category even though Category is not on
the template.
For a database that does not support OLAP functions (in this case Microsoft

Report 6A (MicroStrategy OLAP Function <BreakBy>) - Microsoft Access

select a11.[SUBCAT_ID] AS SUBCAT_ID,
max(a12.[SUBCAT_DESC]) AS SUBCAT_DESC,
a12.[CATEGORY_ID] AS CATEGORY_ID,
from [CITY_SUBCATEG_SLS] a11,
[LU_SUBCATEG] a12
where a11.[SUBCAT_ID] = a12.[SUBCAT_ID]
group by a11.[SUBCAT_ID],
a12.[CATEGORY_ID]
[An Analytical SQL]
If the database supports computation of OLAP functions (in this example

Oracle), then the following SQL is generated:
Report 6B (DB OLAP Function) - Oracle

select a12.SUBCAT_ID SUBCAT_ID,
a13.SUBCAT_DESC SUBCAT_DESC,
a13.CATEGORY_ID CATEGORY_ID,
sum((a11.QTY_SOLD * (a11.UNIT_PRICE - a11.DISCOUNT)))
WJXBFS1,
avg(sum((a11.QTY_SOLD * (a11.UNIT_PRICE -
a11.DISCOUNT)))) over(partition by a13.CATEGORY_ID
order by sum((a11.QTY_SOLD * (a11.UNIT_PRICE -
a11.DISCOUNT))) asc rows 4 preceding) WJXBFS2
from ORDER_DETAIL a11,
LU_ITEM a12,
LU_SUBCATEG a13
where a11.ITEM_ID = a12.ITEM_ID and
a12.SUBCAT_ID = a13.SUBCAT_ID
group by a12.SUBCAT_ID,
a13.SUBCAT_DESC,
a13.CATEGORY_ID

Sorting in OLAP Functions

For OLAP functions, sorting is done before performing computations.
Basically, there are two choices for the SortBy parameter setting:
l Sort by Value in Subexpression: See Reports 5B (in Report 5B (DB

OLAP Function - Window Size) - Oracle 9i, page 71) and 6B (in Report 6B
(DB OLAP Function) - Oracle , page 73) for examples. Note that if the
database can perform computation of OLAP functions, then the definition
of the subexpression M1A displays in the order by clause.
For example, the above reports contain: "order by sum((a11.QTY_

SOLD * (a11.UNIT_PRICE - a11.DISCOUNT))) asc" to represent
the <SortBy = Value ascending> parameter setting.
l Sort by Objects: This type of sorting can use either attributes or metrics.
The following examples demonstrate the differences in processing when
an attribute or metric is used.
In this example, the OLAP metric is sorted by a normal attribute (either by ID

or Desc).
Create an OLAP metric defined as:
OM2 = RunningSum<BreakBy = {[Customer Region]}, SortBy =

([Customer City]@ID asc, Customer@Name dsc)>([M1A])
Then add the attributes Customer Region and Customer, and the metric OM2
to the template.
The metric OM2 is sorted by Customer City@ID in ascending order, then by

Customer@Name in descending order, though the attribute Customer City is
not on the template. Assume that attribute Customer is a child of the
attribute Customer City, and Customer City is a child of Customer Region.
For a database that does not support OLAP functions (in this case Microsoft

Report 7A (MicroStrategy OLAP Function <SortBy>) - Microsoft Access
select a14.[CUST_REGION_ID] AS CUST_REGION_ID,

a15.[CUST_REGION_NAME] AS CUST_REGION_NAME,
a11.[CUSTOMER_ID] AS CUSTOMER_ID,
a12.[CUST_LAST_NAME] AS CUST_LAST_NAME,
a12.[CUST_FIRST_NAME] AS CUST_FIRST_NAME,
from [CUSTOMER_SLS] a11,
[LU_CUSTOMER] a12,
[LU_CUST_CITY] a13,
[LU_CUST_STATE] a14,
[LU_CUST_REGION] a15
where a11.[CUSTOMER_ID] = a12.[CUSTOMER_ID] and
a12.[CUST_CITY_ID] = a13.[CUST_CITY_ID] and
a13.[CUST_STATE_ID] = a14.[CUST_STATE_ID] and
a14.[CUST_REGION_ID] = a15.[CUST_REGION_ID]
[An Analytical SQL]

Oracle 9i), the following SQL is generated:
Report 7B (DB OLAP Function <SortBy>) - Oracle

select a15.CUST_REGION_ID CUST_REGION_ID,
a16.CUST_REGION_NAME CUST_REGION_NAME,
a12.CUSTOMER_ID CUSTOMER_ID,
a13.CUST_LAST_NAME CUST_LAST_NAME,
a13.CUST_FIRST_NAME CUST_FIRST_NAME,
a13.CUST_CITY_ID CUST_CITY_ID,
sum(sum(a11.ORDER_AMT)) over(partition by
a15.CUST_REGION_ID order by a13.CUST_CITY_ID asc,
a13.CUST_LAST_NAME desc, a13.CUST_FIRST_NAME desc rows
unbounded preceding) WJXBFS1
from ORDER_FACT a11,
LU_ORDER a12,
LU_CUSTOMER a13,

LU_CUST_CITY a14,
LU_CUST_STATE a15,
LU_CUST_REGION a16
where a11.ORDER_ID = a12.ORDER_ID and
a12.CUSTOMER_ID = a13.CUSTOMER_ID and
a13.CUST_CITY_ID = a14.CUST_CITY_ID and
a14.CUST_STATE_ID = a15.CUST_STATE_ID and
a15.CUST_REGION_ID = a16.CUST_REGION_ID
group by a15.CUST_REGION_ID,
a16.CUST_REGION_NAME,
a12.CUSTOMER_ID,
a13.CUST_LAST_NAME,
a13.CUST_FIRST_NAME,
a13.CUST_CITY_ID
In both SQLs, the Intelligence Server selects a13.CUST_CITY_ID, even

though attribute Customer City is not on the report. In Report 7B, this
attribute is also in the group by clause. Sort by attribute Customer City@ID
is done because of attribute Customer that is a child of Customer City.
In Report 7B, the SQL has to issue group by CUST_LAST_NAME and

CUST_FIRST_NAME due to the sort by Customer@Name. For optimization
purposes, the Intelligence Server only groups by the ID column.
There is a VLDB setting that allows you to group by non-ID columns. This
VLDB setting can be set for any report that uses an OLAP metric with a sort
by attribute description.
In this example, an OLAP metric is sorted by an attribute and a metric.
Create an OLAP metric defined as:
OM3 = RunningSum<BreakBy = {[Customer Region]}, SortBy =

(Customer@[Last Name] desc, [M1A] asc)>([M1A])
Then add attributes Customer Region and Customer, and metric OM3 to the
template. The OLAP metric OM3 is sorted by Customer Last Name and then
by the metric [M1A].

For databases that do not support OLAP functions (in this case Microsoft
Report 8A (MicroStrategy OLAP Function <SortBy>) - Microsoft Access

a15.[CUST_REGION_NAME] AS CUST_REGION_NAME,
[LU_CUSTOMER] a12,
[LU_CUST_CITY] a13,
a14.[CUST_REGION_ID] = a15.[CUST_REGION_ID] and
a14.[CUST_REGION_ID] in (3, 7)
[An Analytical SQL]

Oracle), the following SQL is generated:
Report 8B (DB OLAP Function <SortBy>) - Oracle
select a15.CUST_REGION_ID CUST_REGION_ID,

a16.CUST_REGION_NAME CUST_REGION_NAME,
a12.CUSTOMER_ID CUSTOMER_ID,
a13.CUST_LAST_NAME CUST_LAST_NAME,
a13.CUST_FIRST_NAME CUST_FIRST_NAME,
sum(a11.ORDER_AMT) WJXBFS1,
sum(sum(a11.ORDER_AMT)) over(partition by
a15.CUST_REGION_ID order by a13.CUST_LAST_NAME desc,

sum(a11.ORDER_AMT) desc rows unbounded preceding)

WJXBFS2
from ORDER_FACT a11,
LU_ORDER a12,
LU_CUSTOMER a13,
LU_CUST_CITY a14,
LU_CUST_STATE a15,
LU_CUST_REGION a16
where a11.ORDER_ID = a12.ORDER_ID and
a12.CUSTOMER_ID = a13.CUSTOMER_ID and
a13.CUST_CITY_ID = a14.CUST_CITY_ID and
a14.CUST_STATE_ID = a15.CUST_STATE_ID and
a15.CUST_REGION_ID = a16.CUST_REGION_ID
and a15.CUST_REGION_ID in (7, 3)
group by a15.CUST_REGION_ID,
a16.CUST_REGION_NAME,
a12.CUSTOMER_ID,
a13.CUST_LAST_NAME,
a13.CUST_FIRST_NAME
NULL Handling in OLAP Functions

When sorted by Intelligence Server, NULL is placed at the end of a list.
Databases also put NULL at the end, as required by the ANSI standard.
For RunningCount or MovingCount functions, NULL is always ignored. For

other computations, such as RunningSum or MovingAvg, NULL is treated as
zero for the sum of values across the function parameters. This behavior is
consistent for Intelligence Server and ANSI-compliant database OLAP
functions.
Tie Handling in OLAP Functions

Tie Handling is related to the SortBy parameter when you sort a tie in a
metric. Intelligence Server uses a Merge-Sort algorithm, which always
preserves the original order retrieved from the database (via ODBC) to the
Intelligence Server.

It is not known whether databases use the same sort algorithm; therefore,
the Intelligence Server and database computation of OLAP functions can
produce different results whenever there is a tie on the data.
Using Custom Plug-In Functions

Function Plug-Ins in the MicroStrategy Engine allow users to define their
own collection of functions, and then let the Engine use them for further
analysis. The plug-in functions, usually called user-defined functions,
behave as if they are an integrated part of MicroStrategy and are
indistinguishable from all other MicroStrategy functions or operators, such
as Sum, Average, Min, Max, Count, -, +, /, or *. The Intelligence Server
performs standard computations such as Sum, Average, +, /, and so on that
are usually calculated by the database.
MicroStrategy is equipped with approximately 250 functions and operators,

including predefined plug-in functions created by MicroStrategy. They are
intended to provide three full-featured libraries of functions that are most
commonly used by customers, which includes financial functions, math
functions, and statistical functions. Examples include Accrint, IRR, NPV,
Abs, Ln, Log, AvgDev, HomoscedasticTTest, Confidence, and so on. These
functions are located at ...Schema Objects/Functions and
Operators/Plug-In Packages.
For detailed information on individual functions, see Chapter 4, Plug-In

Package Functions and the Function Plug-in Wizard online help.
Creating User-Defined Plug-In Functions

Using the MicroStrategy Function Plug-In Wizard, you can define custom
functions relevant to your business case scenarios. You can create
individual functions or even entire function packages, such as the financial,
mathematical, and statistical packages, provided by MicroStrategy. Guided
by the wizard, you start by creating a Microsoft Visual C++ project with
placeholders where you can add custom analytic code. After the new plug-in
function is created, you need to launch MicroStrategy Developer to import it

so it can be used for all the reports. As mentioned previously, once a

function is imported, it will be used in the same way as any other standard
MicroStrategy function.
For instructions on creating a plug-in function, see the Function Plug-in

Wizard online help.
The main stages of the creation process are described as follows:
l Designing: determines how to implement the analytical procedures into a

computer algorithm.
l Creating: creates the Microsoft Visual C++ project, which is used to build
a library containing your algorithms.
l Implementing: creates the code that embodies the algorithms and

compiling this code into a library that is used by MicroStrategy.
l Importing: adds the library to a MicroStrategy project so that its

algorithms are available for use in the project.
l Executing: creates new metrics that use the algorithms and using those
metrics in a MicroStrategy report.
You can create the following types of functions:
l Single-value functions
l Group-value functions
l OLAP functions
The datatypes of input arguments can be:
l Numeric
l Date
l String
A plug-in function can contain more than one parameter. Supported

datatypes for parameters include:

l Byte
l Short
l Long
l Float
l Double
l String (BSTR)
l Bool
l Date
The following two examples illustrate how plug-in functions could be

defined:
l FORECAST<n>(numeric_vector1) = numeric1
l Aggregate function with a numeric vector as an input argument and a numeric scalar
as an output argument.
l One parameter, n, which is a long integer.
l Given a series of values, and given that those values are each separated by one time
interval, this function predicts what the value will be after n additional time intervals.
For example, you include the unit cost of an item in order to predict how
much the unit cost will be two years later.
FORECAST<2>(UnitCost)
In this example, UnitCost includes a four year history of data, providing

the following yearly values to the FORECAST function.
FORECAST<2>(100,120,140,160)
This results in a value of 200, which is the predicted unit cost two years
from the time of the most recently recorded cost data.

l CUSTOMNUMBERFORMAT(numeric1) = string1
l Simple function with numeric scalar as input argument and string scalar as output
argument.
l No parameters.
l Transforms a number into a string representation of that number not supported by

any of MicroStrategy's out-of-the-box metric format strings.
Example: CUSTOMNUMBERFORMAT(123456789) = '1 2345.6789'
To Install the Function Plug-In Wizard
1. From ...\MicroStrategy\Workstation, click FPWizard.exe.
2. Follow the prompts to install.
To Activate the Function Plug-In Wizard
1. From the Start button, select Programs > Microsoft Visual Studio >
Microsoft Visual C++.
2. From the File menu, select New.
3. On the Projects tab, select MicroStrategy Function Plug-in Wizard.
4. Enter a name in the Project Name text box.
Do not use spaces in the project name. Because the project name will
become the name of the library, use something short and descriptive.
For illustration purposes, the project name myFP is used throughout
the online help.
5. Change the location of the project, if needed. Then click OK. The
Function Plug-in Wizard - Step 1 of 2 window is displayed.

To Access the Function Plug-in Wizard Online Help from the

MicroStrategy Online Help
1. In the master MicroStrategy online help, search for custom functions,

or open the Creating and Modifying Additional Report Objects
folder, expand the Metrics folder, and then expand the Using custom
functions folder.
2. Launch the Using custom functions topic. This topic contains a link to
the Function Plug-in Wizard online help.
3. Click the Function Plug-in Wizard on-line help link. The Function
Plug-in Wizard online help opens in a new help window.
Additional Examples of Functions in Expressions

Hypothesis Testing Example
The reports and report objects in this example can be found in the following
folder: MicroStrategy Tutorial\Public Objects\Reports\
MicroStrategy Platform Capabilities\Advanced Analytics\
Statistics and Forecasting\Hypothesis Testing
Which Call Centers had a Statistically Significant

Increase in the Average Daily Sales in Recent Years?
To answer this question, you must first find the average of daily sales for
recent years for each call center. You accomplish this using the Average
Daily Sales metric, defined as:
Average Daily Sales = Avg(Sum(Revenue) {~+, Day+} ) {~+}
This is a simple metric that contains the nested group-value functions Sum
and Avg.

Next, create a report that uses this metric, placing the Call Center attribute
on the row axis and the Year attribute on the column axis. A report with this
definition is shown below.
In this report it looks like all call centers had a significant increase in the
average daily sales in each successive year. However, an average value by
itself does not represent the complete picture.
To get more information, you can check the standard deviation of daily sales
between each successive year for each call center. You can accomplish this
by adding Standard Deviation in Daily Sales to the report. The metric is
defined as:
StDev Daily Sales = Stdev(Sum(Revenue) {~+, Day+} ) {~+}
In statistics, standard deviation is a value which shows how widely a set of

values differs from the mean.
The resulting report provides the average and the standard deviation of daily
sales for all years, for each call center.

This report shows that all call centers have greater average daily sales in
each successive year. Additionally, the standard deviation of daily sales is
higher for each successive year, with a number of exceptions. A couple of
these exceptions include Atlanta between 2009 and 2010, as well as San
Francisco between 2008 and 2009. The higher standard deviation values
mean that there is more volatility in these results. Therefore, the higher
average daily sales in each successive year are not necessarily a
progressive increase over previous years; they may have been caused by
random fluctuations.
Based on this observation, you next need to find out which call centers have
a statistically significant difference in their average daily sales between
these successive years. You can get the result by testing the hypothesis that
the average daily sales are the same, against the hypothesis that the
average daily sales are significantly different. This must be tested for each
call center.
The hypothesis testing is accomplished by computing the p-value. In

statistics, p-value is the probability of making a decision to reject a fact,
given that the fact is correct. In the context of this example, it is the

probability of making a mistake by concluding that the average daily sales in

2008 is significantly different from 2009, given that actually they are about
the same. In general, you want to restrict this type of error so that it is
smaller than a certain tolerance level. This tolerance level is usually set to
between 2.5% to 10.0%.
If you assume that the standard deviations of daily sales for all years are the
same, then you can use a function called HomoscedasticTTest to compute
the p-value. If you assume that the daily standard deviations are different,
then you must use the function HeteroscedasticTTest. The following
examples consider both of these assumptions and compute the p-value
using HomoscedasticTTest and HeteroscedasticTTest.
This report uses simple metrics based on other metrics, because to

calculate the p-value, you need to compute the daily sales for each call
center for each day. You also need to group the fact Revenue at the levels of
Day and Call Center for all years. The fact table DAY_CTR_SLS is available
at the levels of Employee, Order, and Day.
Call Center is related to the fact table via Employee.
You need a table with a structure similar to the following:
Daily Sales Daily Sales

Call Center Day
(last year) (current year)
1 123456 123456
Northwest ... 123456 123456
N 123456 123456
1 123456 123456
Southeast ... 123456 123456
N 123456 123456

You need to create a temporary table with this structure using metrics.
These metrics are then used in simple metrics to calculate the p-value.
Follow the steps below to build the necessary metrics for the Hypothesis
Testing report.
1. Build a metric to create the column for [Daily Sales (current year)] in
the temporary table. This metric is defined as:
Daily Sales (current year)= Sum(Revenue) {~+, Day+} <

[Current year in sample DB]; @2; ->
The default level notation {~+} is placed on the metric since the default
is to group by an attribute on the report. In this example, that report
attribute is Call Center. The metric condition Current year in
sample DB is also applied to calculate the data for the current year
available in the MicroStrategy Tutorial data.
2. Build a metric to create the column for [Daily Sales (last year)] in the
temporary table. This metric is defined as:
Daily Sales (last year) = Sum(Revenue) {~+, Day+} <

[Last year in sample DB]; @2; -> | [Last Year's] |
This metric uses a metric condition and a transformation.
l The metric condition Last year in sample DB is applied to

calculate the data for the previous year available in the MicroStrategy
Tutorial data.
l The transformation Last Year's is necessary because attribute

Day is a child of attribute Year. To show Daily Sales in Day 1, 2, …, N
of last year together with Daily Sales in Day 1, 2, …, N of the current
year, you must use the transformation to supply this data.
3. Build the simple metrics to compute the p-value for each call center by
using the metrics above. The simple metrics are defined as follows:

HeteroscedasticTTest([Daily Sales (last year)], [Daily

Sales (current year)]) {~+}
HeteroscedasticTTest([Daily Sales (last year)], [Daily

Sales (current year)]) {~+}
Create a report with the two metrics shown above (for p-value) and the
attribute Call Center. The Hypothesis Testing report is shown below.
The resulting report, Hypothesis Testing, is shown above. All call

centers, except for Atlanta, Salt Lake City, Seattle, and Fargo, have a
p-value of less that 5%. This indicates that the probability of making an
error in concluding that the sales have significantly increased is low for
all but these four call centers.
Additionally, even Salt Lake City and Fargo are at 20% or lower. This is
strong evidence that average daily sales for the majority of the call
centers can be attributed to a steady increase, rather than random
fluctuations.

The SQL generated for the report Hypothesis Testing is displayed

below, along with a summary of the actions taken at the end of the SQL
statement:
Pass0 - Execution Duration:0:00:00.06

create table ZZTH0FDXLS4MD000 (
DAY_DATE TIMESTAMP,
CALL_CTR_ID SHORT,
WJXBFS1 DOUBLE)
Pass1 - Execution Duration:0:00:00.34
insert into ZZTH0FDXLS4MD000
select a12.[DAY_DATE] AS DAY_DATE,
from [DAY_CTR_SLS] a11,
[LU_DAY] a12
where a11.[DAY_DATE] = a12.[LY_DAY_DATE]
and a12.[YEAR_ID] in (2010)
Pass2 - Execution Duration: 0:00:00.00
create table ZZTU1RY0YSKMD001 (
DAY_DATE TIMESTAMP,
CALL_CTR_ID SHORT,
WJXBFS1 DOUBLE)
insert into ZZTU1RY0YSKMD001
select a11.[DAY_DATE] AS DAY_DATE,
from [DAY_CTR_SLS] a11,
[LU_DAY] a12
where a11.[DAY_DATE] = a12.[DAY_DATE]
select pa11.[CALL_CTR_ID] AS CALL_CTR_ID,
a13.[CENTER_NAME] AS CENTER_NAME,
pa11.[DAY_DATE] AS DAY_DATE,
pa11.[WJXBFS1] AS WJXBFS1,

from [ZZTH0FDXLS4MD000] pa11,

[ZZTU1RY0YSKMD001] pa12,
[LU_CALL_CTR] a13
where pa11.[CALL_CTR_ID] = pa12.[CALL_CTR_ID] and
pa11.[DAY_DATE] = pa12.[DAY_DATE] and
pa11.[CALL_CTR_ID] = a13.[CALL_CTR_ID]
[Analytical SQL calculated by the Analytical Engine:
select CALL_CTR_ID,
CENTER_NAME,
HomoscedasticTTest(WJXBFS1, WJXBFS2),
HeteroscedasticTTest(WJXBFS1, WJXBFS2)
from [previous pass]
]
[Populate Report Data]
drop table ZZTH0FDXLS4MD000
drop table ZZTU1RY0YSKMD001
l Pass0 and Pass1 are issued by Intelligence Server to compute the metric
Daily Sales (last year). The Intelligence Server prepares the temporary
table with Call Center and Day as its key. Then, it retrieves last year's
sales using the transformation Last Year's.
l Pass2 and Pass3 are issued to compute the metric Daily Sales (current
year). Attributes Call Center and Day are used as keys to the temporary
table. This pass is similar to Pass0 and Pass1 with the key difference
being that the metric does not have a transformation.
l In Pass4 and Pass5, the Intelligence Server computes the p-value for each
call center, using the HeteroscedasticTTest and the HomoscedasticTTest
functions.
l The remaining passes perform final report preparation and drop the
temporary tables.

Confidence Level Example

The reports and report components in this example can be found in the
following folder:
MicroStrategy Tutorial\Public Objects\Reports\

Statistics and Forecasting\Confidence Level
Confidence level is used to determine valuable customers in two slightly

different ways:
l Who are my Valuable Customers? (Example 1), page 91: Determines

valuable customers as those whose average spending is above an upper
bound of sales orders.
l Who are my Valuable Customers? (Example 2), page 97: Determines

valuable customers as those whose average spending is above an upper
bound of sales orders. Additionally, valuable customers are compared for
each Customer Region, rather than comparing customers in all regions.
Who are my Valuable Customers? (Example 1)

The basic goal is to define a cut-off value that represents the minimum
requirement to be classified as a valuable customer. To identify the valuable
customers in your customer base, you must determine the parameters that
help differentiate those customers from the others. In this example, valuable
customers are those whose average spending is above an upper bound of
sales orders.
The ORDER_FACT table contains all orders received. Assume that the sales
order amount is normally distributed with a certain mean and standard
deviation. Based on the assumption of normal distribution and a confidence
level of 99%, you can define valuable customers as those who have a
spending average above an upper bound of sales orders.
To determine the valuable customers based on this criteria, you need to

know several values. Use the following metrics to obtain these values.

1. You must know the average sales order from all orders in the ORDER_
FACT table. This is the arithmetic mean of the normal distribution. You
can get this value using the Avg function. The metric to compute the
average of sales orders from the ORDER_FACT table can be defined as:
M01 = Avg(Revenue) {![Call Center]+, !Year+,

!Employee+, !Order+, !Day+}
Since Day, Employee, and Order are the highest attributes that are
parents of the keys in the ORDER_FACT table, include them as the level
of aggregation to make sure that the ORDER_FACT table is used. Also,
the average must be calculated over all sales order amounts, so you
must set the group-by on the level attributes to none. This ensures that
the metric does not group by any of these attributes.
2. You must know the standard deviation of sales orders from all orders.
This number is the standard deviation of the normal distribution. You
can get this value using the StDevP function (see StDevP (Standard
Deviation of a Population), page 140).
The metric to compute the population standard deviation of sales

orders from the ORDER_FACT table is defined as follows:
M02 = StdevP(Revenue) {![Call Center]+, !Year+,

3. You must know the number of rows in the ORDER_FACT table. You need
this value to calculate the cut-off value. You can retrieve this value by
using the Count function, in particular, Count from the ORDER_FACT
table. The metric to count the data in the ORDER_FACT table is defined
as follows:
M03 = Count(Revenue) {![Call Center]+, !Year+,
4. Under the assumption of normal distribution with the parameters given

above, you must determine the cut-off value that represents the top 5%

of sales order amounts. To do this, you can use the Confidence plug-in
function. The metric to compute the cut-off number is defined as
follows, using some of the metrics defined above:
M04 = Confidence(0.01, [M02], [M03])
The number 0.01 comes from a normal distribution with 99%

confidence level.
5. The metric that calculates the upper bound uses some of the metrics
defined above, and is defined as follows:
M05 = ([M01] + [M04])
6. You must calculate the average sales for each customer to get a list of
valuable customers based on the criteria. The metric is defined as
follows:
Average Sales = Avg(Revenue) {~+}
You do not need to include the level of aggregation
for Average Sales since the default {~+} notation is
replaced by the attribute Customer.
7. You must also apply a metric qualification in a filter to restrict the rows
returned to those that meet the definition of valuable customers. The
filter is defined as follows:
F01 = Set of Customers where (Average Sales Greater

than [M05]
To generate a report listing valuable customers, create a report with the

attribute Customer on the row axis and the attribute Customer Region
on the page-by axis. Apply filter F01 to the report to produce a list of
valuable customers for each attribute element in Customer Region. The
Valuable Customers 01 report is shown below:

The SQL generated for the report Valuable Customers 01 is as follows:

create table ZZTLZV82TJ0MD000 (
CUSTOMER_ID LONG,
WJXBFS1 DOUBLE)
insert into ZZTLZV82TJ0MD000
select a11.[CUSTOMER_ID] AS CUSTOMER_ID,
avg(a11.[TOT_DOLLAR_SALES]) AS WJXBFS1
from [CUSTOMER_SLS] a11
group by a11.[CUSTOMER_ID]
a11.[ORDER_ID] AS ORDER_ID,
a11.[EMP_ID] AS EMP_ID,

a11.[ORDER_DATE] AS DAY_DATE,
a11.[CUSTOMER_ID] AS CUSTOMER_ID0,
(a11.[QTY_SOLD] * (a11.[UNIT_PRICE] - a11.[DISCOUNT]))
AS WJXBFS1
from [ORDER_DETAIL] a11
create table ZZTW4A12VECMD001 (
CUSTOMER_ID LONG,
WJXBFS1 DOUBLE)
select CUSTOMER_ID0,
(ISNULL(avg(WJXBFS1), 0) + Confidence(0.01,
StdevP(WJXBFS1), count(WJXBFS1)))
]
insert into ZZTW4A12VECMD001 values (1, 33.4351602545)
create table ZZTPDKUSLWSMQ002 (
CUSTOMER_ID LONG)
insert into ZZTPDKUSLWSMQ002
select pa11.[CUSTOMER_ID] AS CUSTOMER_ID
from [ZZTLZV82TJ0MD000] pa11,
[ZZTW4A12VECMD001] pa12
where pa11.[CUSTOMER_ID] = pa12.[CUSTOMER_ID]
and (pa11.[WJXBFS1] > pa12.[WJXBFS1])
max(a13.[CUST_LAST_NAME]) AS CUST_LAST_NAME,
max(a13.[CUST_FIRST_NAME]) AS CUST_FIRST_NAME,
a15.[CUST_REGION_ID] AS CUST_REGION_ID,
max(a16.[CUST_REGION_NAME]) AS CUST_REGION_NAME0,
avg(a11.[TOT_DOLLAR_SALES]) AS WJXBFS1
[ZZTPDKUSLWSMQ002] pa12,

[LU_CUSTOMER] a13,
[LU_CUST_CITY] a14,
where a11.[CUSTOMER_ID] = pa12.[CUSTOMER_ID] and
a11.[CUSTOMER_ID] = a13.[CUSTOMER_ID] and
group by a11.[CUSTOMER_ID],
a15.[CUST_REGION_ID]
drop table ZZTLZV82TJ0MD000
drop table ZZTW4A12VECMD001
drop table ZZTPDKUSLWSMQ002
l Pass0 and Pass1 compute the average sales per customer (Average Sales
metric) and put the results in a temporary table.
l Pass2 retrieves ORDER_DETAIL data, puts the data into memory, and uses
the data to compute M01, M02, and M03.
l Pass3 creates a temporary table used to store the results of the

calculations.
l Pass4 calculates M02 using the Intelligence Server because the group-
value function StDevP is not supported by the Access database. The
values of M01, M02, and M03 are used to compute M05.
l Pass5 inserts the results of the calculations into the temporary table.
l Pass6 and Pass7 use the earlier two temporary tables to qualify on
valuable customers based on the definition.
l Pass8 displays the report with attributes Customer and Customer Region.

l The remaining passes perform additional report preparation and drop the
temporary tables.
Who are my Valuable Customers? (Example 2)

This example generates a list of valuable customers based on a different
definition from the previous example.
Assume a normal distribution of sales orders is still valid, but the cut-off
value is for each Customer Region. This means you must compare the
average spending of each customer against the average spending of the
Customer Region to which the customer belongs. To allow for more
variation, take the average value from the current year data, but use sample
standard deviations for all available years. Use a confidence level of 95%.
Valuable customers are defined as those customers who have total spending
above the cut-off value. The cut-off value is calculated using the filters and
metrics defined below:
1. The filter used to limit the calculation of average sales orders to only
the data for the current year is defined as follows:
Current year in sample DB = Year In list (2010)
2. The metric used to calculate the average sales order is defined as

follows:
M07= Avg(Revenue) {[Customer Region]+, !Year+,

!Employee+, !Order+, !Day+} <[Current year in sample
DB]; @2; ->
3. The metric used to calculate the standard deviation of sales orders is

defined as follows:
M08 = Stdev(Revenue) {[Customer Region]+, ![Call

Center]+, !Year+, !Employee+, !Order+, !Day+}

Notice that the function Stdev (standard deviation of a sample) is used

in this metric because the amount of data is limited to each Customer
Region. This means the data can be considered as more of a sample
than a full population.
4. The metric used to determine the cut-off value is defined as follows,

using the preceding metrics:
M09 = ([M07] + (1.96 * [M08]))
The number 1.96 comes from a normal distribution with 95%

confidence level. In practice, this number is often rounded to 2 instead
of 1.96.
5. The metric used to determine each customer's spending so that it can

be compared to the cut-off value is defined as follows:
M10 = Sum(Revenue){~+}
6. The metric qualification (filter) needed to restrict the report to a list of

valuable customers based on Definition 2 is defined as follows:
F03 = Set of Customer where (M10 Greater than [M09])
7. To view the last order sales from all valuable customers, use the metric
defined as follows:
Last Order Sales = Sum(Revenue) {~+, >|Day+}
You must compute order sales for each customer at the Day level from
the ORDER_FACT table, but must take it from the last Day on that fact
table. The {~+} is replaced by any attribute on the report. In this
example, it is the Customer attribute.
Build a report putting the attribute Customer on the row axis, the
attribute Customer Region on the page-by axis, and the Last Order
Sales metric on the column axis, then applying the filter F03. The
Valuable Customers 02 report is shown below.

The SQL generated by the Valuable Customers 02 report is as follows:

create table ZZT3WSP7T8BMD000 (
CUSTOMER_ID LONG,
WJXBFS1 DOUBLE)
insert into ZZT3WSP7T8BMD000
create table ZZT36UF785VMD001 (
CUST_REGION_ID SHORT,

WJXBFS1 DOUBLE)
insert into ZZT36UF785VMD001
avg((a11.[QTY_SOLD] * (a11.[UNIT_PRICE] -
a11.[DISCOUNT]))) AS WJXBFS1
from [ORDER_DETAIL] a11,
[LU_CUSTOMER] a12,
[LU_CUST_CITY] a13,
[LU_DAY] a15
a11.[ORDER_DATE] = a15.[DAY_DATE]
group by a14.[CUST_REGION_ID]
a11.[ORDER_ID] AS ORDER_ID,
a11.[EMP_ID] AS EMP_ID,
a11.[ORDER_DATE] AS DAY_DATE,
(a11.[QTY_SOLD] * (a11.[UNIT_PRICE] - a11.[DISCOUNT]))
AS WJXBFS1
[LU_CUSTOMER] a12,
[LU_CUST_CITY] a13,
[LU_CUST_STATE] a14
a13.[CUST_STATE_ID] = a14.[CUST_STATE_ID]
create table ZZTS6MZ3N2BMD002 (
CUST_REGION_ID SHORT,
WJXBFS1 DOUBLE)


select CUST_REGION_ID,
ISNULL((1.96 * Stdev(WJXBFS1)), 0)
]
insert into ZZTS6MZ3N2BMD002 values (1, 127.4088118401)
create table ZZTSMWZF9SRMQ003 (
CUSTOMER_ID LONG)
insert into ZZTSMWZF9SRMQ003
select distinct pa11.[CUSTOMER_ID] AS CUSTOMER_ID
from [ZZT3WSP7T8BMD000] pa11,
[LU_CUSTOMER] a12,
[LU_CUST_CITY] a13,
[ZZT36UF785VMD001] pa15,
[ZZTS6MZ3N2BMD002] pa16
where pa11.[CUSTOMER_ID] = a12.[CUSTOMER_ID] and
a14.[CUST_REGION_ID] = pa15.[CUST_REGION_ID] and
pa15.[CUST_REGION_ID] = pa16.[CUST_REGION_ID]
and (pa11.[WJXBFS1] > (IIF(ISNULL(pa15.[WJXBFS1]), 0,
pa15.[WJXBFS1]) + IIF(ISNULL(pa16.[WJXBFS1]), 0,
pa16.[WJXBFS1])))
create table ZZTRX7EEWHNNB004 (
DAY_DATE TIMESTAMP,
CUSTOMER_ID LONG,
WJXBFS1 DOUBLE)
insert into ZZTRX7EEWHNNB004
select a11.[ORDER_DATE] AS DAY_DATE,

sum((a11.[QTY_SOLD] * (a11.[UNIT_PRICE] -
a11.[DISCOUNT]))) AS WJXBFS1
[ZZTSMWZF9SRMQ003] pa12
where a11.[CUSTOMER_ID] = pa12.[CUSTOMER_ID]
group by a11.[ORDER_DATE],
a11.[CUSTOMER_ID]
create table ZZTQFB0EMCJMB005 (
CUSTOMER_ID LONG,
WJXBFS1 TIMESTAMP)
insert into ZZTQFB0EMCJMB005
select pc11.[CUSTOMER_ID] AS CUSTOMER_ID,
max(pc11.[DAY_DATE]) AS WJXBFS1
from [ZZTRX7EEWHNNB004] pc11
group by pc11.[CUSTOMER_ID]
select distinct pa11.[CUSTOMER_ID] AS CUSTOMER_ID,
a16.[CUST_REGION_NAME] AS CUST_REGION_NAME0,
from [ZZTRX7EEWHNNB004] pa11,
[ZZTQFB0EMCJMB005] pa12,
[LU_CUSTOMER] a13,
[LU_CUST_CITY] a14,
where pa11.[CUSTOMER_ID] = pa12.[CUSTOMER_ID] and
pa11.[DAY_DATE] = pa12.[WJXBFS1] and
pa11.[CUSTOMER_ID] = a13.[CUSTOMER_ID] and


drop table ZZT3WSP7T8BMD000
drop table ZZT36UF785VMD001
drop table ZZTS6MZ3N2BMD002
drop table ZZTSMWZF9SRMQ003
drop table ZZTRX7EEWHNNB004
drop table ZZTQFB0EMCJMB005
l Pass0 and Pass1 compute metric M10.
l Pass2 and Pass3 compute metric M07.
l Pass4, Pass5, Pass6, and Pass7 compute (1.96 * M08). The group-value
function for M08, StDev, is not supported by the database (Microsoft
Access) so it is calculated by Intelligence Server.
l Pass8 and Pass9 obtain the list of valuable customers, evaluating the filter
condition F03.
l Pass10, Pass11, Pass12, and Pass13 compute the Last Order Sales
metric.
l Pass 14 displays the results for the report.
l All other passes drop the temporary tables.
Statistical Descriptors - Simple Example

The report and report components used in this example can be found in the
MicroStrategy Tutorial project in the following folder:
Statistics and Forecasting\Statistical Descriptors

The Component Objects - Advanced folder contains additional

statistical descriptor examples.
How Do I Learn More About My Customers?

You can get more information about the customers in each region by
performing statistical calculations on their spending. This example uses
analysis to answer the following questions:
l How many customers exist in each customer region?
l What is the average spending of customers in each customer region?
l What is the median spending of customers in each customer region?
l What is the standard deviation of customer spending in each customer

region?
l How much is the 25th and 75th percentile of customer spending in each
customer region?
These are statistical descriptors about customers in each customer region.

This type of data can help you understand how customers spend their
money.
To answer the questions above, you must create a set of simple metrics
based on other metrics. These metrics are defined using nested group-value
functions.
1. This metric determines the number of customers in each region:

Count of Customers = Sum(Count(1) {~+, Customer+} )
{~+}
2. This metric determines the average spending of customers in each

region:
a.Mean Revenue Per Customer = Avg(Sum(Revenue) {~+,
Customer+} ) {~+}

3. This metric determines the median spending of customers in each

region:
b.Median Revenue Per Customer = Median(Sum(Revenue)
{~+, Customer+} ) {~+}
4. This metric determines the standard deviation of customer spending in

each region:
e.StDev of Revenue Per Customer = Stdev(Sum(Revenue)

{~+, Customer+} ) {~+}
5. This metric determines the threshold for the 25th percentile:

25th Percentile of Revenue Per Customer = Percentile
(Sum(Revenue) {~+, Customer+} , 0.25) {~+}
6. This metric determines the threshold for the 75th percentile:
75th Percentile of Revenue Per Customer = Percentile

(Sum(Revenue) {~+, Customer+} , 0.75) {~+}
Create a report and place all of the metrics described above on the row
axis and place the attribute Customer Region on the column axis.
Execute the Statistical Descriptors - Simple report, which is shown
below:
This report provides information on the behavior of customer spending

in each customer region. For example, the Northwest customer region
has the least number of customers, but most customers in that region
spend more than any other region, as seen in the larger mean and
median revenues per customer. This can highlight that it is worthwhile
to gain a larger customer base in this region.

However, the somewhat lower value for the standard deviation of

revenue per customer in the Northwest region also indicates that this
result might be due to the smaller sample size and a few customers that
spend abnormally large amounts. This standard deviation value helps
to show that while a campaign to increase the customer base in the
region is still an attractive idea, the results should be monitored to
determine if revenues continue to be higher per customer.
The SQL generated for the Statistical Descriptors - Simple report is as

follows:

create table ZZT42W71C95MD000 (
CUSTOMER_ID LONG,
WJXBFS1 LONG,
WJXBFS2 DOUBLE)
insert into ZZT42W71C95MD000
count(1.0) AS WJXBFS1,
select distinct pa11.[CUSTOMER_ID] AS CUSTOMER_ID,
a15.[CUST_REGION_NAME] AS CUST_REGION_NAME0,
from [ZZT42W71C95MD000] pa11,
[LU_CUSTOMER] a12,
[LU_CUST_CITY] a13,
where pa11.[CUSTOMER_ID] = a12.[CUSTOMER_ID] and

select CUST_REGION_ID,
CUST_REGION_NAME0,
sum(WJXBFS1),
avg(WJXBFS2),
Median(WJXBFS2),
Stdev(WJXBFS2),
Percentile(WJXBFS2, 0.25),
Percentile(WJXBFS2, 0.75)
]
drop table ZZT42W71C95MD000
l The first two passes of SQL (Pass0 and Pass1) prepare and calculate the
values used in other metrics.
l The Intelligence Server optimizes the process by pulling only a single

column of revenue data even though it is used in several other metrics.
l In Pass2 and Pass3 the Intelligence Server retrieves the results from
Pass1 to compute the other metrics. In Pass3 the Intelligence Server acts
as an in-memory database, using the values from Pass2 to calculate the
following group-value functions: Sum, Avg, Median, Stdev, and Percentile.
l The remaining passes prepare the report and drop the temporary tables.
Forecasting Example
The report and report components in this example can be found in the
MicroStrategy Tutorial project in the following folder:
Statistics and Forecasting\Forecasting

How to Forecast Future Sales Based on Existing Sales Data?

The MicroStrategy Tutorial project has sales data for multiple years. Using
this data, you can forecast potential sales for the upcoming years.
To plot a sales line for the expected future sales data, apply linear
extrapolation techniques to the historical data. Linear extrapolation involves
the assumption that the trend of past data will continue in a linear fashion.
The slope and Y-intercept values are calculated based on historical data,
and the same slope and Y-intercept are applied to extend the data into the
future.
This is just one type of forecasting analysis that can be done in

MicroStrategy. There are additional MicroStrategy functions that provide
other variations on calculating a forecast of values. For information on these
functions, how they forecast values, and forecasting examples using these
functions, see:
l ForecastV (Forecast, Vector Input), page 453
l GrowthV (Growth, Vector Input), page 459
l TrendV (Trend, Vector Input), page 488
To perform the analysis, build a set of metrics that calculate each of the
components in the analysis, then use those metrics together to create the
final Forecasting metric.
1. The metric that calculates sales revenue at the Year level is defined as
follows:
Revenue {Year} = Sum(Revenue) {~+, Year+}
2. The metric that expresses the Year elements as a series of integers, to

simplify the analysis, is defined as follows:
YearNumber = RunningSum<SortBy= (Year@ID) >(Sum(1)
{~+, Year } )

3. The metric that calculates the slope of the line produced by the
Revenue {Year} metric is defined as follows:
SlopeMetric= Slope([Revenue {Year}], YearNumber) {~+,

!Year+}
4. The metric that calculates the Y-intercept of the line produced by the
Revenue {Year} metric is defined as follows:
InterceptMetric = Intercept([Revenue {Year}],

YearNumber) {~+, !Year+}
5. The final metric combines the three preceding metrics into a forecast
value. This formula follows the standard formula for a line: y = mx +
b. The metric is defined as follows:
Forecast = ((SlopeMetric * YearNumber) +

InterceptMetric)
Place the Forecast metric on a report with Year. A sales forecast is

generated for all available years. The Forecast (Graph) report displays
as follows:

The sales for 2008, 2009, and 2010 are actual recorded values, and the
sales for 2011 are a linear extrapolation of the existing data into the
future.
The following SQL is generated for the Forecast (Graph) report:

create table ZZT16JMSID1MD000 (
YEAR_ID SHORT,
CATEGORY_ID SHORT,
WJXBFS1 DOUBLE)
insert into ZZT16JMSID1MD000
select a11.[YEAR_ID] AS YEAR_ID,
from [YR_CATEGORY_SLS] a11
select a12.[YEAR_ID] AS YEAR_ID,
1.0 AS WJXBFS1
from [LU_CATEGORY] a11,
[LU_YEAR] a12
create table ZZT3M2QSNWDMD001 (
YEAR_ID SHORT,
CATEGORY_ID SHORT,
WJXBFS1 DOUBLE)
select YEAR_ID,
CATEGORY_ID,
RunningSum<SortBy= ([YEAR_ID])>(WJXBFS1)
]
insert into ZZT3M2QSNWDMD001 values (2008, 1, 1)

create table ZZTDMBESMXPOJ002 (

YEAR_ID SHORT,
CATEGORY_ID SHORT)
insert into ZZTDMBESMXPOJ002
select pa11.[YEAR_ID] AS YEAR_ID,
pa11.[CATEGORY_ID] AS CATEGORY_ID
from [ZZT16JMSID1MD000] pa11
insert into ZZTDMBESMXPOJ002
from [ZZT3M2QSNWDMD001] pa11
create table ZZT9MCAB4STOD003 (
YEAR_ID SHORT,
CATEGORY_ID SHORT)
insert into ZZT9MCAB4STOD003
select distinct pa11.[YEAR_ID] AS YEAR_ID,
from [ZZTDMBESMXPOJ002] pa11
pa11.[CATEGORY_ID] AS CATEGORY_ID,
from [ZZT9MCAB4STOD003] pa11,
[ZZT16JMSID1MD000] pa12,
[ZZT3M2QSNWDMD001] pa13
where pa11.[CATEGORY_ID] = pa12.[CATEGORY_ID] and
pa11.[YEAR_ID] = pa12.[YEAR_ID] and
pa11.[CATEGORY_ID] = pa13.[CATEGORY_ID] and
pa11.[YEAR_ID] = pa13.[YEAR_ID]
create table ZZTMYCAAR65MD004 (
CATEGORY_ID SHORT,

WJXBFS1 DOUBLE,
WJXBFS2 DOUBLE)
select CATEGORY_ID,
Slope(WJXBFS1, WJXBFS2),
Intercept(WJXBFS1, WJXBFS2)
]
insert into ZZTMYCAAR65MD004 values (1,
58938.0812500116, 585341.077083335)
select pa11.[CATEGORY_ID] AS CATEGORY_ID,
a13.[CATEGORY_DESC] AS CATEGORY_DESC0,
pa11.[YEAR_ID] AS YEAR_ID,
(IIF(ISNULL((pa12.[WJXBFS1] * pa11.[WJXBFS1])), 0,
(pa12.[WJXBFS1] * pa11.[WJXBFS1])) +
IIF(ISNULL(pa12.[WJXBFS2]), 0, pa12.[WJXBFS2])) AS
WJXBFS1
from [ZZT3M2QSNWDMD001] pa11,
[ZZTMYCAAR65MD004] pa12,
[LU_CATEGORY] a13
where pa11.[CATEGORY_ID] = pa12.[CATEGORY_ID] and
pa11.[CATEGORY_ID] = a13.[CATEGORY_ID]
drop table ZZT16JMSID1MD000
drop table ZZT3M2QSNWDMD001
drop table ZZTDMBESMXPOJ002
drop table ZZT9MCAB4STOD003
drop table ZZTMYCAAR65MD004

l Pass0 and Pass1 calculate the Revenue {Year} metric.
l Pass2 calculates the inner portion of the YearNumber metric, assigning

the integer 1 to each year available in the LU_YEAR lookup table. Note that
this metric was defined by adding Year dimensionality with filtering set to
None. This forces the metric into its own pass of SQL, to ensure that all
years in the lookup table are numbered, not just the years with data in the
fact table.
l Passes 3 to 14 calculate the metrics YearNumber, SlopeMetric, and

InterceptMetric. The Intelligence Server performs the calculations based
on the data in the temporary table from Pass0, Pass1, and Pass2. Values
are calculated and inserted back into the temporary tables.
l In Pass15, the Forecast metric is calculated based on the information from

all of the previous calculations. The mx+b calculation is performed as part
of the SQL statement.
l The remaining passes prepare the report and drop the temporary tables.

STAN D ARD FUN CTION S

The standard functions in this chapter range from simple operations like Sum
and Product to internal functions that allow you to define your own metrics,
facts, and filters. Taken as a set, they provide a powerful toolset for
performing common mathematical calculations, grouping data, examining
correlation, validating data, and formatting report information for display.
The following categories of functions are covered:
l Basic Functions, page 116
l Date and Time Functions, page 149
l Internal Functions, page 169
l NULL/Zero Functions, page 188
l OLAP Functions, page 191
l Rank and NTile Functions, page 268
l String Functions, page 293
Each section in this chapter describes a function category and presents a

list of functions within each category. Each section also presents the data
necessary to understand and implement each individual function. The
information provided for each function includes
l An explanation of the data returned by the function
l The syntax of the function including function name, available parameters,

the parameter setting defaults, and the types of data that can be used with
the function
l The mathematical expression illustrating exactly how the calculation is

defined in MicroStrategy (if applicable)
l Usage notes describing any error conditions, invalid data types, or key
information to know before using the function (if applicable)
l An example of the function in use; this may be either a report example or a

simple text description of the data returned based on the specified input

l The report examples in this section use objects and data found in the
MicroStrategy Tutorial.
l See Appendix A, MicroStrategy and Database Support for Functions, for a

list of the databases and the functions they support.
Basic Functions
These are basic mathematical functions like Avg, Greatest, Least, Max, Min,
and so on, and are among the most commonly used functions.
Add
Returns the sum of two or more values. The Add function differs from the
Sum function (see Sum , page 144), which returns the sum of values in a
single value list. Add can take multiple metrics as inputs and add the values
of the metrics.
You can also construct these types of metrics using the plus operator (see
Plus (+), page 322) instead of the Add function.
Syntax
Add(Argument1, Argument2,..., ArgumentN)
Where
The arguments must be metrics or constants.
Example
You can use the Add function to return the sum of related metrics. In the
report shown below, the Total Paid Compensation metric uses the formula
Add([Paid Bonus],[Paid Salary]) to return the total compensation
paid to each employee. This report was created in the MicroStrategy Human
Resources Analysis Module.

Average
Average performs a sum of a set of values and divides this total by the
number of values in the set. This returns the average (also known as
arithmetic mean) for the set of values.
The Average function differs from the Avg function (see Avg (Average),
page 117), which returns the average of values in a single value list. The
Average function can take multiple metrics as inputs and average the
values of the metrics. You can use this function to compute and display the
average of two or more metrics on a report.
Syntax
Average(Argument1, Argument2,..., ArgumentN)
Where:
Avg (Average)
Avg calculates the sum of a single value list and divides the result by the
number of values in the list. This returns the average (arithmetic mean) of
the listed values.

Avg is often used to create subtotals and metrics based on fact data. This is
a group-value function.
To calculate an average while applying a weight to the values, see

WeightedMean, page 260.
Syntax
Avg<Distinct, FactID
Where:
l Argument is an attribute, fact, or metric representing a list of numbers.
l Distinct is a TRUE/FALSE parameter that allows you to use all values in

the calculation or to calculate using only the unique values.
l FactID is a parameter that forces a calculation to take place on a fact

table that contains the selected fact.
Usage Notes
The Avg function ignores NULL values but uses zero values in its
calculation.
Example
Example 1: In this simple example the average of a value list containing the
values (4,9,2,9) is calculated as follows:
(4+9+2+9)/4 = 6
Example 2: This report obtains the average salary for employees in each
region. The report contains the attributes Region, Employee, and Salary as
well as the metric Average Salary. A report filter limits the regions displayed
to Northeast and Southeast. The metric Average Salary is defined as
follows: Avg(Salary){Region}
The function sums all salaries within a region and divides by the number of
values, resulting in the regional average salary.

For further examples of using the Avg function, see Example 1: Transformed
Fact, page 12 and Example 2: Compound Metric, page 13, as well as
Example 1: Average, page 14.
Condition
Condition is a shortcut function, available for various features in
MicroStrategy Web, that allows you to easily define the condition (filtering)
of the final metric expression. A conditional metric allows you to apply a
filter to only one metric on a report without affecting the other metrics. The
metric filter can be either a filter or a prompt that returns a list of filters to
choose from. Only one filter or prompt can be associated with each metric,
but the metric filter can contain multiple qualifications. For additional
information on when to create conditional metrics, see the Advanced
Reporting Guide.
Once you select the Condition shortcut function, you can then select the
group function for the calculation, such as Sum or Max. Additionally, the
options to add or remove a condition are displayed:
l To add a condition to the metric, click the Browse icon ( ) to select a

filter or a prompt that returns a list of filters.
l To remove the condition from a metric, click the Delete icon ( ).
You can also define how the condition is evaluated by clicking Condition
advanced options:

l By default, the parts of the report filter that are based on parent or child
attributes of the attribute in the metric condition are ignored, and do not
affect the calculation of the conditional metric. To apply all criteria in the
report filter to the conditional metric, clear the Ignore related report filter
elements check box. This option is selected by default.
l By default, the report filter is applied to the metric data and then the metric
filter is applied to those results. You can determine the order in which the
report filter and metric filter are applied to the metric. From the
Interaction between metric filter and report filter drop-down list, select
one of the following:
l To apply the report filter criteria first, then apply the metric filter to the
results, select Merge report filter into metric (default). For example,
the metric filter is revenue greater than $100 and the report filter is
bottom 10 items for sales. In this example, the report filter narrows the
result set to only 10 items, and then the metric condition filters out, from
those 10 items, any items with a revenue above $100.
l To evaluate the metric filter first, then apply the report filter to the
results, select Merge metric condition into report. Using this option
with the example above, the metric condition returns all items with a
revenue above $100. The report filter then filters out all but the bottom
10 of those items, based on sales.
l To combine the metric filter and the report filter, select Merge into new.
Only those results that meet both the metric filter and the report filter are
included in the metric. Using this option with the example above, the two
filters are merged, so that only those items that are in the bottom 10 in
terms of sales and that have sales greater than $100 are included.
Count
Count returns the number of elements in a list of values. This is a group-
value function.
Syntax

Count<Distinct, FactID, UseLookupForAttributes, Null>

(Argument)
Where:
l Argument is an attribute, fact, or metric representing a list of values.
l Distinct is a TRUE/FALSE parameter that allows you to count all

elements in a list or only the unique elements.

l UseLookupForAttributes is a TRUE/FALSE parameter that can be

used when performing an aggregation of an attribute:
l If you set this parameter to TRUE, the aggregation is done for the unique
set of attribute elements, excluding any duplicates caused by additional
attributes on the report.
l If you set this parameter to FALSE, the aggregation is done for all
elements of an attribute, including duplicate elements that can be
included by displaying additional attributes on a report.
For an example of how to use this parameter, see Performing Counts of

Attributes, page 122.
Example
This example creates a report that calculates the count of employees in

each region. The report contains the attributes Region and Employee, and
the metric # of Employees. A report filter limits the regions displayed to
Central, South, Southeast, and Southwest. The metric # of Employees is
defined as follows: Count(Employee){Region, ~}.
The function counts each entry within a region and returns the last number in
the count.

Performing Counts of Attributes

While most functions are commonly used with facts and metrics, the Count
function is commonly used to count the number of elements for an attribute.
For example, the report shown below displays the customer revenue for
each item they purchased.

This report also uses two derived metrics to display the following count
information:
l Number of Customers: This derived metric displays the count of

customers. The expression used to define the derived metric is
Count<UseLookupForAttributes=True>(Customer){}. By defining
the parameter UseLookupForAttributes=True, the Count function
ignores any multiple listings of each Customer attribute element. In the
report shown above, each separate item is ignored and only the unique
attribute elements for Customer are counted.
l Number of Customer Purchases: This derived metric displays the count

of customer purchases. The expression used to define the derived metric
is Count<UseLookupForAttributes=False>(Customer){}. By
defining the parameter UseLookupForAttributes=False, the Count
function counts each listing of the Customer attribute element. In the
report shown above, each separate item is counted, which provides a
count of customer purchases.
For information on creating derived metrics, along with other OLAP Services
features, see the In-memory Analytics Guide.
First
Returns the result of an aggregate applied over a set of rows that ranks as
the first within a specified order. This is a group-value function that shares
the sort by capability of the OLAP functions.
Syntax
First<FactID, SortBy>(Argument)
Where:


l SortBy is a parameter that allows you to define the attribute or metric on

which to sort the data.
Example
The First function is often used to define subtotals. In this example, the First
function defines the subtotal for the metric units received, returning the units
received for the first quarter containing a value in 2003. The expression
defines the subtotal as follows:
First<SortBy= (Day, Month, Quarter, Year) >([Units Received]){@}
Argument is the metric to which the subtotal is applied.
For detailed instructions on creating and applying user-defined subtotals,

see the MicroStrategy Advanced Reporting Guide.

GeoMean (Geometric Mean)

This function can be used to find the average for a set of values contributing
to a product. While the Avg function is used to find the arithmetic mean for a
set of values contributing to a sum, GeoMean can be used to determine the
average growth rate for a given compound interest with variable rates.
This function takes the product of a set of values of size N and returns the
Nth root of the result (also known as geometric mean). GeoMean is a group-
value function.
Syntax
GeoMean <FactID>(Argument)
Where:

Expression
Usage Notes
This function returns an error if a value in the value list is negative.
Example
Suppose you have an investment that earns 10% the first year, 60% the
second year, and 20% the third year, and you want to answer the question
"What is its average rate of return?" You do not want to use the Avg function
to obtain the arithmetic mean, because these numbers show that in the first
year your investment was multiplied by 1.10, in the second year by 1.60, and
in the third year by 1.20. The average is the geometric mean of these three
numbers. This can be understood as looking for a constant that you can

multiply each year's investment by, and get the same result as multiplying
the first year by 1.10, the second year by 1.60, and so forth. This constant is
the geometric mean.
(1.10 * 1.60 * 1.20) 1/3 = 1.283
The geometric mean is 1.283, so the average rate of return is about 28%.
Greatest
Returns the larger of two or more values. The Greatest function differs
from the Max function, which returns the largest value in a single value list.
Greatest can take multiple lists as input and compare the elements in the
lists. It is used for comparisons between metrics.
Syntax
Greatest(Argument1, Argument2,..., ArgumentN)
Where:
The arguments must be metrics.
Example
This simple example illustrates how data is returned by the Greatest

function.
Given two value lists, the function compares the values in each position in
the list and returns a list containing the highest numbers from each position.
Result
List 1 List 2
List
21 50 50
18 3 18
42 22 42
30 6 30
7 20 20

Histogram Median
Histogram Median is a distributive implementation to calculate
approximation of median over large datasets.
Formula
Where:
L m is the lower limit of the median bar.
n is the total number of observations in the set.
F m-1 is the total number of observations in all bars below the median bar.
f m is the frequency of the median bar.
c is the median bar width.
Syntax
HistogramMedian([ValueList], NumberOfHistograms)
Where:
ValueList is an attribute, fact, or metric representing a list of values.
NumberOfHistograms represents the histograms to be analyzed.
Example
This example shows a report to find the median Revenue per Customer for
each Income Bracket in the Central Region.
HistogramMedian([Max Revenue per Customer], [Max Revenue

per Customer]) {~ , Region}

Last
The Last function returns the last value in a sorted set of values. It is often
used to define subtotals, as shown in the example below. This is a group-
value function.
Syntax
Last<FactID, SortBy>(Argument)
Where:

l SortBy is a parameter that allows you to define the attribute or metric on

which to sort the data.
Example
In this example, Last is used in the subtotal for the metric End on Hand,
returning the total items on hand in the last quarter of 2003. The subtotal
expression is defined as follows:
Last<SortBy= (Day, Month, Quarter, Year) >(End on Hand) {@ }
For detailed instructions on creating and applying user-defined subtotals,

see the MicroStrategy Advanced Reporting Guide.

Least
This function returns the smaller of two or more values. The Least function
differs from the Min function, which returns the smallest value in a single
value list. Least can take multiple lists as input and compare the elements
of the lists. It is used for comparisons between metrics. This is a single-
value function.
Syntax
Least(Argument1, Argument2,..., ArgumentN)
Where:
The arguments must be metrics.
Example
This simple example illustrates how data is returned by the Least function.

Given two values lists, the function compares the values at each position in
the lists and returns a list of the lowest values from each position.
Result
List 1 List 2
List
21 50 21
18 3 3
42 22 22
30 6 6
7 20 7
Level
Level is a shortcut function, available for various features in MicroStrategy
Web, that allows you to easily define the level (dimensionality) of the final
metric expression. The level of a metric determines the conceptual level at
which a calculation is performed. For example, rather than just calculating
your revenue, you want to specify that a metric always calculates yearly
revenue. For additional information on when to create level metrics, see the
Advanced Reporting Guide.
Once you select the Level shortcut function, you can then select the group
function for the calculation, such as Sum or Max. Additionally, the options to
add and remove levels are displayed:
l Report Level is the default level. If you remove Report Level as a level of
the metric, you can add it back by typing Report Level in the level text
field.
l To add a level to the metric, click the Browse icon ( ) to select an

attribute.
l To remove a level from a metric, click the Delete icon ( ).
You can also define how levels are evaluated:

l To define the level options, for each level, click the Level options icon (
).
Drop-down List Level Options
To include only data that meets the conditions in the report filter in
the metric calculation, select Standard filtering.
To raise the level of the report filter to the level of the target, if
possible, then apply the report filter to the metric calculation, select
Absolute filtering. For example, the report filter contains the
Washington, DC, Boston, and New York call centers, but the
Revenue metric is calculated at the Region level. Because Call
Center is a child attribute of Region, the report filter's level is raised
From the to the Region level, and the report filter is treated as if it includes the
Relationship with regions that contain Washington, DC, Boston, and New York (in this
Report Filter drop- case, Mid-Atlantic and Northeast). Data from all call centers in the
down list, you can Mid-Atlantic or Northeast regions are included in the metric
define how the calculation, including call centers that are not Washington, DC,
report filter affects Boston, or New York.
the metric
calculation. To ignore any conditions in the report filter that are based on the
target attribute, as well as any parent or child attributes of the
target, select Ignore filtering. For example, if you have a regional
revenue metric on a report, any conditions based on Country,
Region, or Call Center in the report filter are ignored when
calculating the metric. However, a condition based on Year would
not be ignored, since Year is not directly related to Region.
To allow the target and group components of the level to define the
filter, select None.
From the Metric

Aggregations
drop-down list, you
can determine how To group data in the metric by the attribute level of the target, select
the metric is Standard.
grouped, or
aggregated, when
displayed on a

Drop-down List Level Options
To exclude the target attribute from being used to group data in the
metric, select None. Any children of the target attribute are also
excluded. This option is available for metrics calculated at a set
level, as opposed to the report level.
The following options are only used for nonaggregatable metrics. A

nonaggregatable metric, such as an inventory metric, is one that
should not be aggregated across an attribute. For example, if you
have monthly inventory numbers in your data warehouse and want
to calculate the yearly inventory, adding the monthly numbers
report. together does not provide a useful business measure. Instead, you
may want to use the end-on-hand and beginning-on-hand inventory
numbers to see how the total inventory changed during the year.
The following options are available:
• To use the first value in the lookup table, select Beginning

lookup.
• To use the last value in the lookup table, select Ending lookup.
• To use the first value in the fact table, select Beginning fact.
• To use the last value in the fact table, select Ending fact.
l To define the advanced level options, such as whether to apply the metric
filter to the metric calculation, click Level advanced options:
l Allow other users to add extra units to this definition: This option is
used to emulate MicroStrategy 6.x behavior and affects only those
projects that have been upgraded from 6.x. Clear the check box only if
your project was upgraded from MicroStrategy 6.x. This option is
selected by default.
l Include filter attributes which are not in report or level in metric

calculation: Determine whether to apply the metric filter when the
metric is calculated. If this option is cleared, filter attributes that are not
on the report or in the level of the metric are not included in the metric

calculation. This options lets you determine which parts of the metric
filter are applied based on the data that has been included on the report.
This option can help you re-use the same metric in multiple reports,
eliminating the need to create and maintain multiple metrics, particularly
if the metric and filter qualifications are complex. For an example,
including an explanation of how the report SQL is affected by this option,
see the Advanced Reporting Guide. This option is selected by default.
Max (maximum)
Max returns the largest value in a list of values. For example, it can be used
on a list of prices to determine the maximum cost of an item. This is a group-
value function.
To compare the highest values in multiple lists of values, use the Greatest
function.
Syntax
Max <FactID
Where:
l Argument is an attribute, fact or metric representing a list of numbers.

Example
This example shows a dossier grid displaying the sales range of items within
a subcategory (Aisle). The Max function is used to determine the highest
sales of an item within a subcategory. The metric, Maximum Sale, is defined
as follows:
Max (Sales){~}

The metric is placed on the grid with the attributes Department and Aisle and
the metrics Sale Total and Minimum Sale. The resulting grid appears as
follows:
Median
Median returns the value in the middle of a set of listed values. The result
will be greater than half the values in the list and less than the other half.
The median can be used as an alternative to the arithmetic mean when
handling values that are not evenly distributed or contain outliers. This is a
group-value function.
Syntax
Median <FactID>(Argument}
Where:


Usage Notes
This function provides a location measure: the value returned has a relative
position with regard to other values in the list. Mode is another function that
provides a location measure.
Example
This example shows a report built to obtain the median age of employees
within each region. The report includes the attributes Region, Employee,
and Employee Age, and the metric Median Age. A report filter limits the
regions displayed to Mid Atlantic, Northeast, and Southeast. The metric
Median Age is defined as follows:
Median([Employee Age){Region,~}
The function evaluates the list of employee ages within a region and selects
a value in the middle of the value list.

For another example of using the Median function, see Example 2: Median,
page 14.
Min (minimum)
Min returns the lowest value in a list of values. For example, it can be used
on a list of prices to determine the minimum cost of an item. This is a group-
value function.
To compare the lowest values in multiple lists of values, use the Least
function.
Syntax
Min <FactID, NullInclude>(Argument)
Where:

l NullInclude is a parameter that determines how NULL values are

included in the calculation.
Example
This example shows a report displaying the cost range of items within a
subcategory. The Min function is used to determine the lowest cost of an
item within a subcategory. The metric Minimum Unit Cost is defined as
follows:
Minimum([Unit Cost]) {~}
The metric is placed on a report with the attributes Category and

Subcategory and the metric Maximum Unit Cost.

Mode
Returns the most frequently occurring value in a given list. This is a group-
value function.
Syntax
Mode <FactID, IsRemovable>(Argument)
Where:

l IsRemovable is a parameter that determines whether the Mode of a

single value returns that value or a NULL value, as described below:

l IsRemovable=True returns the single value as the result of the Mode

function. For example, Mode<IsRemovable=True>(4) would return
the value 4.
l IsRemovable=False returns a NULL value as the results of the Mode

function when a single value is supplied. For example,
Mode<IsRemovable=False>(4) would return NULL.
If multiple values are supplied to the Mode function, the IsRemovable

parameter has no effect on the result of the Mode function.
Usage Notes
l This function provides a location measure: the value returned has a

relative position with regard to other values in the list. Median is another
function that provides a location measure.
l Returns a NULL if there are no recurring values in the range or list.
Example
This example shows a report that retrieves the mode of the unit cost for
items in a subcategory. The report returns the most frequently repeated unit
cost within the various subcategories, and, in the cases of Electronics, TVs,
and Video Equipment, it returns a NULL value, because there are no
recurring unit cost values for them. The report contains the attributes
Category and Subcategory, and the metric Mode Unit Cost. A report filter
limits the categories displayed to Books and Electronics. The Mode Unit
Cost metric is defined as follows:
Mode([Unit Cost])
The formula listed above uses the Unit Cost fact rather than the Unit Cost
metric.

Multiply
Multiply returns the product of two or more values. This function differs
from the Product function (see Product, page 140), which returns the
product of values in a single value list. The Multiply function can take
multiple metrics as inputs and multiply the values of the metrics.
You can also construct these types of metrics using the times operator (see
Times (*), page 321) rather than the Multiply function.
Syntax
Multiply(Argument1, Argument2,..., ArgumentN)
Where:
Example
In the report shown below, the Unit Profit metric uses the formula Multiply
([Unit Price],[Units Sold]) to return the revenue for each item. This
report was created in the MicroStrategy Tutorial project.

Product
Multiplies all values in a list. This is a group-value function.
Syntax
Product<FactID>(Argument)
Where

Example
Given a value list containing the values (1,2,3,4), the product calculates as
1*2*3*4 = 24.
StDevP (Standard Deviation of a Population)

The standard deviation is a value which shows how widely values in a
population differ from the mean. It is useful for comparing different sets of
values with a similar mean.
StDevP returns the standard deviation of a population represented by list of

values. It is a group-value function.
Syntax

StDevP <FactID >(Argument)
Where:

Expression
Usage Notes
l For this function, arguments correspond to an entire population as

opposed to a population sample. For population samples, see StDev
(Standard Deviation of a Sample), page 142.
l When very large population samples are used, this function and the StDev
(Standard Deviation of a Sample), page 142 function return approximately
equal results.
Example
This example shows a report where the standard deviation of the revenue is
calculated. This calculation is based on the assumption that the list of
values supplied in the metric represents the entire population of the data for
which you want to obtain the standard deviation. The calculation is based on
the revenue values for each state within a region and calculated at the
region level.
Compare this example report to the example for StDev to see the different
values returned when calculating for a population as opposed to a sample.
The report contains the attributes Customer Region and Customer State,
and the metrics Total Revenue and StDevP. A report filter limits the regions

displayed to South, Northwest, and Southwest. The definition of the StDevP

metric is as follows:
StDevP([Total Revenue]){[Customer Region], ~}
StDev (Standard Deviation of a Sample)

The standard deviation is an indicator of how widely values in a group differ
from the mean. It is useful for comparing different sets of values with a
similar mean.
StDev returns the standard deviation of a population based on a sample.

This is a group-value function.
Syntax
StDev <Distinct, FactID>(Argument)
Where:

l Distinct is a TRUE/FALSE parameter that allows you to calculate using

all values in the list or only the unique values.

Expression
Usage Notes
l In this function, arguments correspond to a population sample as opposed

to the entire population. For entire populations, see StDevP (Standard
Deviation of a Population), page 140.
l To perform a weighted standard deviation, see WeightedStDev (Weighted

Standard Deviation of a Sample), page 262.
l When very large population samples are used, this function and the
StDevP (Standard Deviation of a Population), page 140function return
approximately equal results.
Example
This example shows a report where the standard deviation of the revenue is
calculated. This calculation is based on the assumption that the list of
values supplied in the metric represents a sample of the data for which you
want to obtain the standard deviation. The calculation is based on the
revenue values for each state within a region and calculated at the region
level.
Compare this example report to the example for StDevP to see the different
values returned when calculating for a population as opposed to a sample.

and the metrics Total Revenue and StDev. A report filter limits the regions
displayed to South, Northwest, and Southwest. The definition of the StDev
metric is as follows:
StDev([Total Revenue]){[Customer Region], ~}
Sum
Returns the sum of all numbers in a list of values. This function is commonly
used in metrics and subtotals. Sum is a group-value function.
Syntax
Sum <Distinct, FactID
Where:



Example
In this example, the metric Regional Revenue is defined as follows:
Sum(Revenue) {[Customer Region], ~}
This simple report uses the attributes Customer Region and Customer State,
and the metrics Revenue and Regional Revenue to generate a report
showing the sum of the revenue for several regions. A report filter limits the
regions displayed to Northeast, Northwest, and Southeast.
Transformation
Transformation is a shortcut function, available for various features in
MicroStrategy Web, that allows you to easily define transformations for the
final metric expression. Transformations allow you to apply an attribute-
element based offset to compare metric data. For example, a transformation
metric can help a user compare last month's revenue to this month's
revenue. Although transformations can be applied to any attribute hierarchy,
the Time hierarchy is used most often. For the Time hierarchy, the offset can
be set as a fixed number of days, weeks, months, or years. For additional

information on when to create transformation metrics, see the Advanced

Reporting Guide.
Once you select the Transformation shortcut function, you can then
select the group function for the calculation, such as Sum or Max.
Additionally, the options to add or remove a transformation are displayed:
l To add a transformation to the metric, click the Browse icon ( ) to select

a transformation.
l To remove a transformation from a metric, click the Delete icon ( ).
VarP (Variance of a Population)

Variance is a measure of how spread out a set of values is.
VarP returns this measure based on an entire population. This is a group-

value function.
Syntax
VarP <FactID>(Argument)
Where:

Expression
Usage Notes

For this function, arguments relate to an entire population as opposed to a

population sample.
Example
Variance is calculated as the average squared deviation of each number

from its mean. This example creates a report in which the variance of the
units sold in the subcategories within a category is calculated with the
assumption that the data provided represents the entire population.
Compare this example report to the example for variance of a sample to see
the different values returned when calculating for a population as opposed
to a sample.
The report includes the attributes Category and Subcategory, and the
metrics Units Sold and VarP by Category. A report filter limits the categories
displayed to Books and Movies. The VarP by Category metric is defined as
follows:
VarP([Units Sold]) {Category, ~}
Var (Variance of a Sample)

Variance is a measure of how spread out a set of values is. It is calculated
as the average squared deviation of each number from its mean.

This function calculates the variance based on a sample of a population.

This is a group-value function.
Syntax
Var <Distinct, FactID>(Argument)
Where:


Expression
Usage Notes
Assume when using this function that arguments constitute a population

sample, as opposed to a entire population.
Example
This example creates a report in which the variance of the units sold in the
subcategories within a category is calculated, with the assumption that the
data provided is a sample.
Compare this example report to the example for variance of a population to

see the different values returned when calculating for a population as
opposed to a sample.

The report includes the attributes Category and Subcategory, and the
metrics Units Sold and Var by Category. A report filter limits the categories
displayed to Books and Movies. The Var by Category metric is defined as
follows:
Var([Units Sold]) {Category, ~}
Date and Time Functions

This section describes the date and time functions. These functions are
supported by the Analytical Engine.
AddDays
AddDays is used to calculate dates that occur N days before or after a given
date. It returns this information in the form of a date or timestamp.
Syntax
AddDays(Date/Time,Offset)
Where:
l Date/Time is the input date or timestamp.
l Offset is an integer number of days to add.

For information on whether your database supports various date and time
functions, see the Appendix A, MicroStrategy and Database Support for
Functions.
Example
AddDays('2020-07-29', 4) = 2020-08-02
AddDays('2020-07-29 02:00:00', 4) = 2020-08-02 02:00:00
AddDays('2020-03-01', -4) = 2020-02-26
The following example can be used to change a TimeDate attribute or metric

to a different time zone. Subtract four hours to change the time zone from
GMT to EST. This does not account for daylight savings time.
AddDays(TimeDate, -4/24)
AddMonths
AddMonths is used to calculate dates that occur N months before or after a
given date. It returns this information in the form of a date or timestamp.
If the new date does not occur in the new month, the last date of the new
month is returned.
Syntax
AddMonths(Date/Time,Offset)
Where:
l Offset is an integer number of months to add.
functions, see Appendix A, MicroStrategy and Database Support for
Functions.
Example
AddMonths('2004-07-29', 3) = 2004-10-29

AddMonths('2004-07-29 10:00:00', 3) = 2004-10-29 10:00:00
AddMonths('2003-03-31', -1) = 2003-02-28
CurrentDate
Returns the current date as provided by the database timer.
This function does not take input variables.
Syntax
CurrentDate()
Functions.
CurrentDateTime
Returns the current date and time as provided by the database timer.
Syntax
CurrentDateTime()
Functions.
CurrentTime
Returns the current time as provided by the database timer.
Syntax
CurrentTime()

Functions.
Date
Returns only the date portion of the date-time column. The time is truncated,
not rounded.
Syntax
Date(Date/Time)
Where:
Date/Time is the input date or timestamp.
Functions.
Example
Input is the date 2004-07-29; output is the date 2004-07-29.
Input is the date-time 2004-07-29 02:00:00; output is the date 2004-

07-29.
DateDiff
DateDiff is used to calculate the length of time between two dates. A
numeric value is returned.
Syntax
DateDiff<firstWeekDay>(Date1, Date2, Unit)
Where:
l Date1 and Date2 are the inputs used for date or timestamp values. You
can use metrics, constants, attribute forms, or functions that result in a

date or timestamp. For example, you can include CurrentDate() as an

input to compare historical dates from an attribute form to the current date.
l Unit is the unit of time that is to be measured. You can return the length
of time in one of the following units:
Unit of Value in the function

time expression
Seconds "s"
Minutes "mn"
Hours "h"
Days "d"
Weeks "w"
Months "m"
Quarters "q"
Years "y"
l firstWeekDay (default value is 1) is a parameter that can be used if you

are returning the length of time in weeks. This parameter determines
which day of the week is considered as the first day of the week, so that
the difference in weeks can be accurately determined. You can type an
integer value from 1 to 7, with 1 representing Sunday, 2 representing
Monday, and so on until 7 representing Saturday.
Functions.
Example
You can use DateDiff to create a metric or attribute form that lists the length
of time between two dates. One common way to do this is to compare the

date information you have stored to the current date. Consider a report or
dossier that includes a Day attribute with a single ID form. You can create a
metric with the following definition:
DateDiff(Day@ID, CurrentDate(), "d")
For each element of the Day attribute, the metric displays the number of
days between that date and the current date when running the report or
dossier.
When using the DateDiff function in MicroStrategy Web, you will need to
replace the Day@ID attribute in the function definition. Create the metric
[MAX(Day@ID)] and define the function as follows: DateDiff([MAX
(Day@ID)], CurrentDate(), "d")
DayofMonth
Returns a number corresponding to the day of the month of the date
provided. The value returned is an integer between 1 and 31.
Syntax
DayofMonth(Argument)
Where:
Argument is a metric representing a list of dates.
Functions.
Example
Input is the date 7/11/2002; output is the integer 11.

DayofWeek
Returns the number of the day in the week corresponding to the input date.
The return value is an integer between 1 and 7. The value of 1 represents
Sunday and the value of 7 represents Saturday.
Syntax
DayofWeek(Argument)
Where:
Functions.
Example
Input is the date 5/16/2003; output is the integer 6, which represents

Friday.
Input is the date 2003/9/9; output is the integer 3, which represents

Tuesday.
DayofYear
Returns the number of the day in the year of the input date. The return value
is an integer between 1 and 365.
Syntax
DayofYear(Argument)
Where:
Functions.

Example
DaysBetween
Returns the difference in days between two given dates as an integer value.
The calculation of the difference is based on the number of day boundaries
crossed, not the number of twenty-four hours periods elapsed. If the first
date argument is later than the second date argument, the result is a
negative number. The result does not display the time stamps, as they are
truncated before performing the calculation.
Syntax
DaysBetween(Date/Time1, Date/Time2)
Where:
l Date/Time1 is the start date.
l Date/Time2 is the end date.
Functions.
Examples
Inputs are the dates 2004-07-29 and 2004-07-25; output is the integer -
4.
Inputs are the dates 2004-07-29 02:00:00 and 2004-07-31 01:00:00;

output is the integer 2.
FiscalMonth
Returns the numeric position of a month within a fiscal year, for a given input
date. This function is useful in financial reporting when the start of the fiscal

year is different than the start of the calendar year.
Syntax
FiscalMonth<firstMonth>(Date/Time)
Where:
l firstMonth (default value is 1) is a parameter that determines which

month is considered as the start of the fiscal year. You can type an integer
value from 1 to 12, with 1 representing January, 2 representing February,
and so on until 12 representing December.
Functions.
Example
Consider a report or dossier that includes a Day attribute with a single ID

form. You can create a metric with the following definition:
FiscalMonth<firstMonth=4>(Day)
For each element of the Day attribute, the metric displays the numeric
position of the month within the fiscal year for that date. For this example,
since the fiscal year starts in April, a date of July 4, 2014 would return 4.
This is because July is the fourth month in the fiscal year.
FiscalQuarter
Returns the numeric position of a quarter within a fiscal year, for a given
input date. This function is useful in financial reporting when the start of the
fiscal year is different than the start of the calendar year.
Syntax
FiscalQuarter<firstMonth>(Date/Time)
Where:


Functions.
Example

FiscalQuarter<firstMonth=7>(Day)
position of the quarter within the fiscal year for that date. For this example,
since the fiscal year starts in July, a date of October 13, 2014 would return
2. This is because October is in the second quarter of the fiscal year.
FiscalWeek
Returns the numeric position of a week within a fiscal year, for a given input
date. This function is useful in financial reporting when the start of the fiscal
year is different than the start of the calendar year.
Syntax
FiscalWeek<firstWeekDay, firstMonth>(Date/Time)
Where:
l firstWeekDay (default value is 1) is a parameter that determines which day of

the week is considered as the first day of the week. You can type an integer value

from 1 to 7, with 1 representing Sunday, 2 representing Monday, and so on until 7

representing Saturday.

Functions.
Example

FiscalWeek<firstWeekDay=1, firstMonth=7>(Day)
position of the week within the fiscal year for that date. For this example,
since the fiscal year starts in July, a date of July 8, 2014 would return 2. This
is because the first fiscal week runs from July 1 st through July 5 th . Then on
July 6 th , the first Sunday of the fiscal year, the second fiscal week starts.
This week includes July 8 th , and so 2 is returned.
FiscalYear
Returns the fiscal year of the input date. This function is useful in financial
reporting when the start of the fiscal year is different than the start of the
calendar year.
When determining the fiscal year, the year returned is the year in which the
fiscal year ends. For example, if a fiscal year runs from March 1, 2014
through April 30, 2015, the fiscal year is 2015.
Syntax
FiscalYear<firstMonth>(Date/Time)

Where:

Functions.
Example

FiscalYear<firstMonth=2>(Day)
For each element of the Day attribute, the metric displays the fiscal year for
that date. A date of July 4, 2013 would have a fiscal year of 2014.
Hour
Returns the integer value for the hour of the input time. The return value is
an integer between 0 and 23.
Syntax
Hour(Argument)
Where:
Argument is a metric representing a list of dates and times.
Functions.
Example

Input is the time 9:45 am; output is the integer 9.
Input is the time 11:10 pm; output is the integer 23.
Millisecond
Returns the integer value for the millisecond of the input time.
Syntax
Millisecond(Argument)
Where:
Functions.
Minute
Returns the integer value for the minute of the input time. The return value is
an integer between 0 and 59.
Syntax
Minute(Argument)
Where:
Functions.
Example
Input is the time 9:45 am; output is the integer 45.
Input is the time 11:10 pm; output is the integer 10.

Month
Returns the number of the month in the year of the input date. The return
value is an integer between 1 and 12.
Syntax
Month(Argument)
Where:
Functions.
Example
MonthEndDate
Returns the date of the last day of the month in which a date or timestamp
occurs.
Syntax
MonthEndDate(Date/Time)
Where:
functions,see Appendix A, MicroStrategy and Database Support for
Functions.
Example

MonthStartDate
Returns the date of the first day of the month in which a date or timestamp
occurs.
Syntax
MonthStartDate(Date/Time)
Where:
Functions.
Example
MonthsBetween
Returns the difference in months between two dates as an integer value.
The difference is calculated by the number of months elapsed and not by the
number of month boundaries crossed. If the first date argument is later than
the second date argument, the result is a negative number. The result does
not display the timestamps, as they are truncated before performing the
calculation.
Syntax
MonthsBetween(Date/Time1, Date/Time2)
Where:
l Date/Time1 is the start date.
l Date/Time2 is the end date.

Functions.
Example
Inputs are the dates 2004-07-29 and 2004-02-28; output is the integer -5.
Inputs are the dates 2004-07-29 and 2004-09-29; output is the integer 2.
Quarter
Returns the number of the quarter in the year of the input date. The return
Syntax
Quarter(Argument)
Where:
Functions.
Example
QuarterStartDate
Returns the date of the first day of the quarter in which a date or timestamp
occurs.
Syntax
QuarterStartDate<firstMonth>(Date/Time)
Where:


months are considered as the first months of each quarter. You can type
an integer value from 1 to 12, with 1 representing January, 2 representing
February, and so on until 12 representing December. Defining one month
automatically defines the first month for the other quarters as well. For
example, if you specify January as the first month, April, July, and October
are also considered as the first month of quarters.
Functions.
Example

form. You can create an attribute form with the following definition:
QuarterStartDate<firstMonth=1>(Day)
For each element of the Day attribute, the new attribute form displays the
start date of the quarter.
Second
Returns the integer value for the second of the input time. The return value
is an integer between 0 and 59.
Syntax
Second(Argument)
Where:
Functions.

Example
Input is the time 9:45:13 am; output is the integer 13.
Input is the time 11:10:47 pm; output is the integer 47.
ToDateTime (Convert String or Number to a Date or

Timestamp)
The ToDateTime function converts a string of characters or a number into a
date or timestamp.
Syntax
ToDateTime<Pattern=null>(Argument)
Where:
l Argument is a fact, metric, column, or constant value that provides the

strings that are converted to numeric values.
l Pattern is a parameter that determines the date or timestamp format

used for the resulting date or timestamp. By default, dates are expected to
be of the format MM/dd/yyyy. When providing a pattern, enclose the
pattern in double quotes (""). For example:
ToDateTime<pattern="MM/dd/yyyy">(Day@ID)
To specify a pattern, you can use custom formatting symbols that conform
to the standards of the International Components for Unicode. For
information about the custom formatting standards and syntax, see
Formatting Dates and Times in the ICU Documentation.
Example
ToDateTime<pattern="MM/dd/yyyy">(Day@ID)
This example would return the data stored for the ID form of Day in a date
format that includes month, day, and year information.

Week
Returns the number of the week in the year of the input date. The return
Syntax
Week(Argument)
Where:
Functions.
Example
WeekStartDate
Returns the date of the first day of the week in which a date or timestamp
occurs.
Syntax
WeekStartDate<firstDay>(Date/Time)
Where:
l firstDay (default value is 1) is a parameter that determines which day of

the week is considered as the first day of the week. You can type an
integer value from 1 to 7, with 1 representing Sunday, 2 representing
Monday, and so on until 7 representing Saturday.

Functions.
Example

form. You can create an attribute form with the following definition:
WeekStartDate<firstDay=1>(Day)
For each element of the Day attribute, the new attribute form displays the
start date of the week.
Year
Returns the year of the input date. The return value is an integer between
1900 and 9999.
Syntax
Year(Argument)
Where:
Functions.
Example
YearEndDate
Returns the date of the last day of the year in which a date or timestamp
occurs.
Syntax

YearEndDate(Date/Time)
Where:
Functions.
Example
YearStartDate
Returns the date of the first day of the year in which a date or timestamp
occurs.
Syntax
YearStartDate(Date/Time)
Where:
Functions.
Examples
Internal Functions
The following are internal function types:

l MicroStrategy Apply, or Pass-through, functions provide access to

functionality that is not standard in MicroStrategy but is available in many
Relational Database Management Systems (RDBMS). These
MicroStrategy functions act as containers for non-standard SQL
expressions passed to your warehouse database. The MicroStrategy
Apply functions are listed below:
l ApplyAgg defines simple metrics or facts using group-value aggregation

functions.
l ApplyLogicdefines custom filters using comparison operators.
l ApplyLogic defines custom filters using logical operators.
l ApplyOLAP defines compound functions using database-specific OLAP

functions such as Rank and RunningSlope.
l ApplySimple uses simple operators and functions like +, -, and * to

perform single-value operations at the database level.
l Banding functions are used to differentiate displayed data on a report.

You can divide data into bands in the following ways:
l Banding distributes the values into bands of equal size.
l BandingC (Banding Count) distributes values into a specified number of

bands.
l BandingP (Banding Points) distributes values into bands based on

specific intervals of values.
l Case functions return specified data based on the evaluation of user-

defined conditions.
l Case evaluates multiple expressions until a condition is determined to

be true, then returns a corresponding value.
l CaseV (Case Vector) evaluates a single metric and returns different

values according to the results.
l Coalesce returns the value of the first non-null argument.
Apply (Pass-Through) Functions

MicroStrategy Apply functions provide access to functions or syntactic
constructs that are not standard in MicroStrategy but are provided by
various Relational Database Management System (RDBMS) platforms.
MicroStrategy strongly advises against using Apply functions when

standard MicroStrategy functions can be used to achieve the same goal,
because using RDBMS functions effectively bypasses the validations and
other benefits of MicroStrategy products. Using Apply functions is
recommended only when corresponding functionality does not exist in
MicroStrategy. When you need to use an Apply function, MicroStrategy
encourages you to submit an enhancement request for inclusion of the
desired feature in a future product release.
To use Apply functions, you must have the Use FreeForm SQL Editor
privilege.
Syntax Common to Apply Functions

While each Apply function has its own syntax, the Apply functions share
several syntactic features:
l The "#n" code in Apply function syntax serves as placeholders for the
MicroStrategy objects being passed to your database. The index for
referencing these objects begins with 0 and increases by 1 for each
successive object passed.
For example, ApplySimple("#0 * #1",[col1],[col2]) indicates

that two items, col1 and col2, referenced as #0 and #1, respectively,
are being passed to your database to be multiplied together (by the
database). If the number of references in an Apply function (for example,

#0, #1, and so forth) exceeds the number of objects passed in that
function, MicroStrategy passes the last available object in place of the
extra reference(s).
For example, ApplySimple("#0 * #1 * #2 * #3",[col1],

[col2]) uses two more references than there are arguments to pass, so
MicroStrategy passes #2 and #3, the extra references, as col2, the last
available object in the list.
l To use # as a character rather than a placeholder, use four # characters in

a row. See the syntax below for an example.
ApplyComparison(UPPER(#0) like
'Z####%', Country@DESC)
The SQL for the function is:
Select a.11[COUNTRY_ID] AS COUNTRY_ID

from [LU_COUNTRY] a11
where upper(a11.[COUNTRY_NAME])
like 'Z#%'
l Do not use form groups for the attribute form expression when using Apply
functions, because form groups are ignored by the Analytical Engine.
Instead, use a single form. For example, instead of using
Customer@Name, where Name is defined to contain Customer's first
name, middle name, and last name, use ID or any other single form.
For more general information on Apply functions as well as an example, see

Apply (Pass-Through) Functions, page 17.
ApplyAgg
The ApplyAgg function is used to define simple metrics or facts by using
database-specific, group-value functions. The ApplyAgg function itself is a
group-value function and accepts facts, attributes, and metrics as input.

Depending on your MicroStrategy product and licensing, this function may

not be available.
Syntax
ApplyAgg()
Accepts facts, attributes, and metrics as input.
All placeholders must begin with #0 and increase in increments of 1.
Example
ApplyAgg("Regrsxx(#0,#1)", [Argument 1], [Argument 2] {~+}
ApplyComparison
ApplyComparison is used to define a filter by populating the WHERE clause
in the SQL passed to your RDBMS, and can take facts, attributes, and
metrics as input.
The ApplyComparison function is used with RDBMS comparison operators

such as >, like, and In.

not be available.
Syntax
ApplyComparison()
Example
ApplyComparison ("#0>#1", Store@ID,2)
For another example of the ApplyComparison function, see Example:

ApplyComparison Used to Check a Prompted Date, page 19.

ApplyLogic
The ApplyLogic function is used to define custom filters. It is used with
logical operators such as AND and OR. ApplyLogic is a logical function.

not be available.
Syntax
ApplyLogic()
Accepts logic (Boolean) values as input.
Example
ApplyLogic("#0 and #1", Year@ID>2003, Month@ID>200301)
ApplyOLAP
OLAP functions are group-value functions that take a set of data as input
and generate a set of data as output, usually reordering the set according to
some criteria.
ApplyOLAP is the MicroStrategy Apply function tool used for OLAP

functionality when you wish to use the native capabilities of your RDBMS. It
is used to define compound metrics via database-specific functions such as
Rank(). ApplyOLAP, like ApplySimple, is used to define metrics but
differs in that it only accepts metrics as input.

not be available.
Syntax
ApplyOLAP()
Accepts only metrics as input.
Example

ApplyOLAP("RunningSlope(#0,#1)", [Metric 1], [Metric 2])
ApplySimple
The ApplySimple function is a single-value function. It is used to insert any
single-value, database-specific functions and simple operators directly into
SQL.
In general, ApplySimple can be used to create the following objects:
l Attribute form
For any Apply function, the attribute form in the arguments should be a
single form—not a form group. The engine ignores any definitions based
on attribute forms.
l Consolidation
l Custom group
l Fact
l Metric
l Subtotal
l Transformation
For information about consolidations, custom groups, metrics, and

subtotals, see the Advanced Reporting Help.

not be available.
Syntax
ApplySimple()
Accepts facts, attributes, and metrics as input.
Examples of Object Creation

Expression
Examples
type
ApplySimple("Datediff(YY,#0,getdate())", [BIRTH_DATE]), in which

[BIRTH_DATE] is an attribute
Attribute
ApplySimple("Months_between(sysdate,#0)",[CURRENT_DT]), in which
[CURRENT_DT] is an attribute
ApplySimple("Greatest(#0,#1,#2)", Metric_1, Metric_2,Metric_3)
Compound ApplySimple(" CASE WHEN #0 between 0 and 100000 THEN 'Low' WHEN
Metric #0 between 100001 and 200000 THEN 'Med' ELSE 'High' END ", Sum
(Revenue){~})
Examples in Custom Expressions
l ApplySimple("Datediff(YY,#0,getdate())", [BIRTH_DATE])
l ApplySimple("Months_between(sysdate,#0)", [CURRENT_DT])
Examples: Incorrect Usage
l ApplySimple("Sum(#0)",[Column 1])
l ApplySimple("Count(#0)",[Column 2])
The two examples immediately above are incorrect and should not be used
in your application because of the following two reasons:
l ApplySimple is a single-value function and therefore can only be used with

single-value functions. Sum and Count are both group-value functions and
therefore should not be used with ApplySimple.
l Sum and Count are both MicroStrategy functions and are not database-
specific; therefore, they should not be used with ApplySimple or any other
Apply functions.

Banding Functions
Banding functions are used to group data on a report so that it is both more
comprehensible and aesthetically pleasing than when it is displayed as one
contiguous list. MicroStrategy provides different banding options for you to
use, depending on how you want to divide your data. The banding functions,
their syntaxes, and examples are listed below.
For information on banding functions in custom groups, see the Advanced

Reporting Help.
Banding
This function maps metric values that fall within a certain range to a
particular integer band value. The range and band values are determined by
the parameter input to the function. For example, if 5,000 is the specified
range, the dollar sales are shown in bands of 0 - 5,000, 5,001 - 10,000,
10,001 - 15,000, and so on. Banding is a single-value function.
Syntax
Banding <HasResidue>(Argument, StartAt, StopAt, Size)
Where:
l Argument is a metric.
l StartAt and StopAt are real numbers specifying the full range of the
values to be placed in bands.
l Size is a real number indicating the width of each band.
l HasResidue is a TRUE/FALSE parameter that defines whether an extra

band is created for all values that do not fall within the StartAt and
StopAt values. Defining HasResidue as TRUE allows the function to
create an additional band which is used to identify all values outside of the
defined band range.
Usage Notes

Values in the list that fall outside of the start and end values set in the
function syntax are assigned a band of 0 (zero) in the report interface.
Example
The following example shows how the Banding function acts on report data.
In this case the total revenue for each city is used to divide the report data
into bands:
l The highest total revenue is 405,367 (New York).
l The lowest revenue is 668 (Cleveland).
The metric function syntax is as follows:
Banding([Total Revenue], 1, 410000, 20000)
l Total Revenue is the metric, defined as Sum(Revenue), representing the

list of values acted on by the Banding function.
l 1 is the value at which banding begins.
l 410,000 is the value at which banding ends.
l 20,000 is the size of the range of values included in each band.
Only part of the resulting report is displayed here.
The result is that 21 bands are created, each encompassing a range of

20,000 dollars of total revenue. To determine the number of bands, 410,000
is divided by 20,000 resulting in 20.5 (rounded to nearest integer, 21). Each
band is given an integer value of between 1 and 21.

BandingC (Banding Count)

BandingC returns metric data grouped into a specified number of bands and
assigns integer values to the resulting bands (for example, a total of 25,000
in dollar sales shown in five equal bands). This is a single-value function.
Syntax
BandingC <HasResidue>(Argument, StartAt, StopAt, BandCount)
Where:
l BandCount is a positive integer indicating the number of bands into which

the total is divided.

l StartAt and StopAt are real numbers specifying the full range of the
values to be placed in bands.

band is created for all values that do not fall within the StartAt and
StopAt values. Defining HasResidue as True allows the function to
defined band range.
Usage Notes
Values that fall outside of the range indicated by the Start and Stop values
are assigned a band of 0 (zero).
Example
The following example shows how the BandingC function acts on report
data. This example uses the same set of values as the example for the
Banding function. Using the BandingC function, you can designate the
number of bands created for the range of values.
BandingC([Total Revenue], 1, 410000, 25)
Where:

list of values acted on by the BandingC function.
l 1 is the value at which banding begins.
l 410,000 is the value at which banding ends.
l 25 is the number of bands into which you want the values divided.
The result is that 20 bands of equal sizes are created. To determine the
range of each band, 410,000 is divided by 25 resulting in bands of 16,400.
Each band is given an integer value of between 1 and 25.

BandingP (Banding Points)

Returns metric data grouped into bands identified by user-specified
boundary point values (for example, 0 - 5,000, 5,000 - 20,000, 20,000 -
30,000). This function assigns integer values to the resulting metric and
allows you to create band intervals of varying widths. This is a single-value
function.
Syntax
BandingP <HasResidue>(Argument, Boundary1, Boundary2,...,

BoundaryN)
Where:

l Boundary1 through BoundaryN are real numbers indicating the cut-off

value for each band. Boundary1 is less than Boundary2, Boundary2 is
less than Boundary3, and so on.

band is created for all values that do not fall within the Boundary1 and
BoundaryN values. Defining HasResidue as TRUE allows the function to
defined band range.
Example
The following example shows how the BandingP function acts on report
data. This example again uses the same data set as in the previous
examples (Banding and BandingC). Using the BandingP function, you can
designate the boundaries for each interval.
BandingP([Total Revenue],
1,20000,40000,60000,80000,100000,410000)
Where:

list of values acted on by the BandingP function.
l All other values specify the boundaries of an interval. There are six
intervals defined as follows:
l Band 1: 1 £ [Total Revenue] < 20000
l Band 2: 20,000 £ [Total Revenue] < 40,000

l Band 6: 100,000 £ [Total Revenue] £ 410,000
Case Functions
Case functions return specified data in a SQL query based on the evaluation
of user-defined conditions. In general, a user specifies a list of conditions
and corresponding return values. When MicroStrategy queries the data
warehouse, the software determines which condition evaluates as true and
then returns the value that the user has specified that corresponds to that
condition. The case functions, their syntaxes, and examples are listed
below.

Case
This function evaluates multiple expressions until a condition is determined
to be true, then returns a corresponding value. If all conditions are false, a
default value is returned. Case can be used for categorizing data based on
multiple conditions. This is a single-value function.

not be available.
Syntax
Case (Condition1, ReturnValue1, Condition2,ReturnValue2,...,

DefaultValue)
Where:
l Condition1 is the first condition to evaluate. The condition can contain

metrics, comparison and logical operators, and constants.
l ReturnValue1 is a constant or metric value to return if the Condition1

condition is TRUE.
l Condition2 is the second condition to evaluate. The condition can

contain metrics, comparison and logical operators, and constants.
l ReturnValue2 is a constant or metric value to return if the Condition2

condition is TRUE.
l ... represents any number of Conditions and Return Values that can be
passed through this function.
l DefaultValue is the information to return if none of the other conditions

are TRUE.
Example
This example generates a report where if the revenue, represented by the

Total Revenue metric, is less than 300,000, the function returns a 0; if the

revenue is less than 600,000, the function returns a 1; if revenue is any

other value, the function returns a 2. The case metric is defined as follows:
Case(([Total Revenue] < 300000), 0, ([Total Revenue] < 600000),

1, 2)
Difference Between Case and If
The If function is very similar to the Case function. Each function takes a
condition as an argument and returns a value depending on whether the
condition is true or not. The Case function can evaluate multiple conditional
arguments, while the If function can only evaluate one condition. However,
the If function can be evaluated by either the SQL Engine or the Analytical
Engine, while the Case function is evaluated only by the Analytical Engine.
CaseV (Case Vector)

CaseV evaluates a single metric and returns different values according to
the results. It can be used to perform transformations on a metric. For
example, if provided a list of values ranging from 1 to 12, CaseV might return
January for a value of 1, February for a value of 2, etc. This is a single-value
function.


not be available.
Syntax
CaseV (Argument, Value1, Result1, Value2, Result2, ...,

DefaultResult)
Where:
l Argument is the metric against which all values are compared.
l Value1 is the first value (constant or metric) to be evaluated.
l Result1 is the information to return (constant or metric) if the Value1

value is equal to the value for the Metric.
l Value2 is the second value (constant or metric) to be evaluated.
l Result2 is the information to return (constant or metric) if the Value2

value is equal to the value for the Metric.
l ... represents any number of Conditions and Return Values that can be
passed through this function.
l DefaultResult is the information to return (constant or metric) if none of

the other values are Equal to the Metric.
Usage Notes
The metric or argument in the CaseV expression is always held as a float.

This means that even if the value is 2, it is held as 2.00000; and therefore 2
and 2.00000 are never considered equal. For this reason, it is best if you
wrap the metric or argument in the integer function, for example, CaseV(int
(M1), 2, A,...).
Example
This simple example generates a report where if the Unit Profit for the item
is 2, the function returns a 200; if the Unit Profit for the item is 3, the
function returns a 300; if Unit Profit is any other value, the function returns a

1000000. Notice how a single metric, Unit Profit, is evaluated against

multiple numeric values. The report contains the attribute Item and the
metrics Unit Profit and CaseV. A report filter limits the items displayed to
those in the subcategory Action. The CaseV metric is defined as follows:
CaseV ([Unit Profit], 2, 200, 3, 300, 1000000)
Coalesce
Returns the value of the first non-null argument. Coalesce can be used to
identify data in tables that may not be fully populated or in metric definitions.

not be available.
Syntax
Coalesce (Argument1, Argument2,..., ArgumentN)
Where:
The arguments for the Coalesce function can be any expression that can be
evaluated as null or not null.

Usage Notes
You can use the Coalesce function in defining a metric, but more often it is
used with the Query Builder feature to support the inclusion of the Coalesce
function in SQL queries. See the example below for more detailed
information.
Example
Your database has two tables T1 and T2 that include the column MONTH_ID
with the format yyyymm. You want to filter on a specific month, but you are
not sure which table has been populated with month data. In the Query
Builder Editor, you can filter your SQL query by creating the condition
Coalesce(T1.MONTH_ID, T2.MONTH_ID) = 200410. The WHERE
clause of the SQL query checks for the first non-null MONTH_ID column and
compares it to the value 200410.
For more information on Query Builder, see the Advanced Reporting Help.
NULL/Zero Functions
The NULL/Zero functions are often used to determine how NULLs and zeros
are displayed on a report. They can also be used as a form of data validation
to ensure meaningful results. For example, an otherwise invalid
mathematical expression such as 5 + NULL can be changed to 5 + 0.
IsNotNull
Returns TRUE if value input is not NULL; otherwise, returns FALSE. This is a
comparison function.

not be available.
Syntax
IsNotNull(Argument)

Where:
Argument is a fact or metric.
Usage Notes
The IsNull and IsNotNull functions work only with the data returned from the
database. For example, if the database data is NULL, the IsNotNull function
returns a FALSE. However, you may see NULL data when you manipulate a
report after its original generation and that manipulation does not
regenerate the data from the database. This can happen when you page by,
for example, and the result includes a NULL.
IsNull
Returns TRUE if the value is NULL; otherwise, returns FALSE. This is a
comparison function.

not be available.
Syntax
IsNull(Argument)
Where:
Usage Notes
The IsNull and IsNotNull functions work only with the data returned from the
database. For example, if the database data is NULL, the IsNotNull function
returns a FALSE. However, you may see NULL data when you manipulate a
report after its original generation and that manipulation does not
regenerate the data from the database. This can happen when you page by,
for example, and the result includes a NULL.

NullToEmpty
Converts a value of NULL to an empty string. If the value is not NULL, the
original value is kept.
Syntax
NullToEmpty(Argument)
Where:
Argument is metric or attribute form.
Example
You can use NullToEmpty when creating derived attributes to convert

NULL values for attribute elements to empty strings. For example, you can
convert NULL values to empty strings so that attribute element values can be
successfully concatenated with functions such as Concat (see Concat
(Concatenate), page 295) or ConcatBlank (see ConcatBlank (Concatenate
Plus Blank Space), page 297).
For steps to create derived attributes, see the MicroStrategy Web Help.
NullToZero
Converts a value of NULL to 0. If the value is not NULL, the original value is
kept.
Syntax
NullToZero(Argument)
Where:
Example
In this simple example, the function NullToZero is applied to a list of values

(12, NULL, 97, 43, NULL). The resulting list is (12, 0, 97, 43, 0). This
function could be applied to a value list before it is acted on by another

function such as Average so that NULL values in a list are included as list
elements and factored into the average as zeros.
ZeroToNull
Converts a value of 0 to NULL. If the value is not 0, the original value is kept.
Syntax
ZeroToNull(Argument)
Where
Example
In this simple example, the function ZeroToNull is applied to a list of values

(12, 0, 97, 43, 0). The resulting list is (12, NULL, 97, 43, NULL). This
function could be applied to a value list before it is acted on by another
function such as Average so that zero values in a list are not included as list
elements and therefore not factored into the average.
OLAP Functions
OLAP functions are also known as relative functions. They take multiple
elements from a list and return a new list of elements. The following applies
to all OLAP functions:
l SortBy is applied before the engine performs the calculation of an OLAP

function.
l In OLAP functions, the SortBy parameter can be either a metric or an

attribute.
Many OLAP functions calculate measures useful for analyzing a set of

values such as the sum, average, and standard deviation. These functions
fall into three groups depending on how they select the window of values on
which they base their calculations. These groups are as follows:

l Running: Functions with a running window include the current value and
all preceding values. For example, given the list (1, 2, 3, 4), RunningSum
returns the sums 1, 3, 6, and 10. This example is illustrated in the table
below. These functions include the word Running in their name, such as
RunningAvg and RunningStDev.
l Moving: Functions with a moving window include the current value and a
fixed number of preceding values. For example, given the list (1,2,3,4) and
a window size of 2, MovingSum returns 1, 3, 5, and 7. This example is
illustrated in the table below. These functions include the word Moving in
their name, such as MovingAvg and MovingStDev.
l OLAP: Functions with flexible windows allow you to set where windows
begin and end in relation to the current value. This feature allows you to
include both preceding and succeeding values in your calculations. For
example, you can use OLAPSum to include one value above and below the
current row. This example is illustrated in the table below. These functions
include the word OLAP, such as OLAPSum and OLAPAvg.
The following table lists a comparison of the example scenarios described

above.
Values RunningSum MovingSum OLAPSum
1 1 (1) 1 (1) 3 (1+2)
2 3 (1+2) 3 (1+2) 6 (1+2+3)
3 6 (1+2+3) 5 (2+3) 9 (2+3+4)
4 10 (1+2+3+4) 7 (3+4) 7 (3+4)
ExpWghMovingAvg (Exponential Weighted Moving Average)

ExpWghMovingAvg allows you to place more or less emphasis on recent
data than on past data within a specified number of rows. It is calculated

within the specified window size and can restart based on an attribute
specified in the function parameters.
Syntax
ExpWghMovingAvg <BreakBy, SortBy> (Argument, WindowSize, Rate)
Where:
l BreakBy is the attribute indicating where the calculation restarts.
l SortBy is the attribute or metric by which the data is sorted.
l Argument is a metric representing a list of numbers.
l WindowSize indicates the number of values to use in each calculation.
l Rate is a positive real number specifying the base weight applied to each
argument value. In the calculation, exponents are sequentially applied to
the rate value. Assign a rate of less than one to give more emphasis to
more recent data; assign a rate of greater than 1 to give greater emphasis
to past data.
Expression
Where:
l k = row number
l y i = metric value at the ith row
l m = window size or the row number, whichever is smaller

l n = number of rows
l w = the base weight applied to each value, which is determined by the

Rate value in the function, as described in the function syntax details
above
Rows with null values are excluded from the calculation.
Example
This example uses small numbers to demonstrate the calculation for the
exponential weighted moving average (ExpWghMovingAvg) function. For
example, you have a list of values (32, 8, 5), with 5 being the most recent
value, and you assign a rate of .5 and a window size of 2.
EWM
Values Calculation
average
32(.5) 0 / (.5) 0 = 32(1)/1 = 32

32 32
8(.5) 0 + 32(.5) 1 / (.5) 0 + (.5) 1 =

8 16
8(1)+ 32(.5) / 1+.5 = 8+16 / 1.5 = 16
5(.5) 0 + 8(.5) 1 / (.5) 0 + (.5) 1 =

5 6
5(1)+ 8(.5) / 1+.5 = 5+4 / 1.5 = 6
As an additional example, the Human Resources Analysis Module project

includes the Division Breakdown report shown below.

A description of how the MovingAvg metric is used on the report is provided

in MovingAvg (Moving Average), page 209. You can also add an exponential
weighted moving average metric to this report to apply more or less
emphasis to older data. For example, you can create another derived metric
named ExpWghMovingAvg with the following expression:
ExpWghMovingAvg<BreakBy={Division}, SortBy=(Quarter)> (Employees,

4.0, 0.5)
The rate of 0.5 changes the moving average to apply more significance, or
weight, to more recent data. This means that the number of Sales employees
during the first quarter (13) has less weight than the number of Sales
employees in the fourth quarter (18). When comparing MovingAvg and
ExpWghMovingAvg for the Sales division, you can see that the
ExpWghMovingAvg is larger. This is because more significance is given to
more recent data, and the recent trend is that employment is on the rise.
This is shown in the report below.

Conversely, you can modify the same ExpWghMovingAvg metric to use the
following expression:
ExpWghMovingAvg<BreakBy={Division}, SortBy= (Quarter) >

(Employees, 4.0, 2)
The rate of 2 changes the moving average to apply more significance, or

weight, to older data. This means that the number of Sales employees
during the first quarter (13) has more weight than the number of Sales
employees in the fourth quarter (18). When comparing MovingAvg and
ExpWghMovingAvg for the Sales division, you can see that the
ExpWghMovingAvg is smaller. This is because more significance is given to
older data, and employment was lower earlier in the year. This is shown in
the report below.

ExpWghRunningAvg (Exponential Weighted Running

Average)
ExpWghRunningAvg allows you compute a running average while placing
more or less emphasis on recent data than on past data. The calculation can
restart based on an attribute specified in the function parameters.
Syntax
ExpWghRunningAvg <BreakBy, SortBy> (Argument, Rate)
Where:
l Rate is a positive real number specifying the base weight applied to each
argument value. In the calculation, exponents are sequentially applied to
the rate value. Assign a rate of less than one to give more emphasis to
more recent data; assign a rate of greater than 1 to give greater emphasis
to past data.
Expression

Where:
l k = row number
l y i = metric value at the ith row
l m = window size or the row number, whichever is smaller
l n = number of rows
l w = the base weight applied to each value, which is determined by the

Rate value in the function, as described in the function syntax details
above
Rows with null values are excluded from the calculation.
Example
This example uses small numbers to demonstrate the calculation for the
exponential weighted running average (ExpWghRunningAvg) function. For
example, you have a list of values (32, 8, 5), with 5 being the most recent
value, and you assign a rate of .5.
EWR
values calculation
average
32(.5) 0 / (.5) 0 = 32(1)/1 = 32

32 32

EWR
values calculation
average
8(.5) 0 + 32(.5) 1 / (.5) 0 + (.5) 1 =

8 16
8(1)+ 32(.5) / 1+.5 = 8+16 / 1.5 = 16
5(.5) 0 + 8(.5) 1 + 32(.5) 2 / (.5) 0 + (.5) 1 +

5 10 (.5) 2 = 5(1)+ 8(.5)+ 32(.25) / 1+ .5 + .25 =
5 + 4 + 8 / 1.75 = 10 (rounded from 9.71)
FirstInRange
Returns the first value in a range of values. FirstInRange can be used to
examine data such as inventory at the beginning of each month.
Calculations can be restarted based on attributes set in the function

parameters. This restart capability differentiates FirstInRange from
First. While First is a group-value function and takes both facts and
metrics as argument input, FirstInRange is an OLAP function and takes
only metrics as argument input.
Syntax
FirstInRange <BreakBy, SortBy> (Argument)
Where:
Expression
Where:

l k = row number
l y 1 = first non-NULL value of the metric after applying the BreakBy and
SortBy parameters
Example
This example shows the results of using the FirstInRange function. The
report includes the attributes Customer Region and Customer State and the
metrics Profit and FirstInRange. A report filter limits the regions displayed to
Northwest, Southeast, and Southwest. The FirstInRange metric is defined as
follows:
FirstInRange<BreakBy={[Customer Region]}, SortBy= ([Customer

State]) >(Profit)
In the following report, FirstInRange returns the first profit value in the list of
Customer States for each Customer Region.
Lag
The Lag function is useful to display a set of metric values in an order
determined by another metric or attribute on the report. This allows you to
compare metric values side-by-side. The easiest way to understand the Lag
function is with an example, provided below.

Syntax
Lag <BreakBy, SortBy> (Argument, Offset, DefaultValue)
Where:
l Argument is an attribute or metric representing a list of values. It is

common to use a metric for Argument.
l Offset specifies the offset from the current row to display information for.
This offset trails behind the current row (you can use the Lead function to
use an offset that precedes ahead of the current row, see Lead, page 207).
For example, with an offset of two, row three displays data for the row two
behind it, which would be row one.
l DefaultValue is the value displayed for any entries that do not have any
data due to the offset. For example, when using an offset of 2, the first two
entries use the default value.
Some common default values to display include:
l 0: To display a value of 0 for any entries that do not have any data due
to the offset, type 0.
l No data: To display no data for any entries that do not have any data due
to the offset, type ZeroToNull(0).
Example
On a report with Item, Cost, and Profit, you can use the Lag function to
create a Cost (Lag on Profit) metric. This metric displays Cost values based
on the descending sort order of the Profit metric, and is defined with the
following expression:
Lag<SortBy= (Profit Desc) >(Cost, 3.0, 0.0)

Notice that the offset is set to three, meaning that the display of cost values
lags behind by three entries. This is displayed in the report results shown
below.
Only the top and the bottom of the report is shown. To view the entire report
results, you can create this report in the MicroStrategy Tutorial project.
The Cost (Lag on Profit) values are displayed three behind the Cost metric,
and are displayed based on the descending sort order of Profit. Notice that
with an offset of three, the first three entries for Cost (Lag on Profit) use the
default value of zero. Also, the last three values of Cost are not included in
the Cost (Lag on Profit) metric.
The report has been sorted based on the Profit metric to make the Cost (Lag
on Profit) values easier to analyze.

Transformation-Style Analysis Using the Lag and Lead Functions

Transformation-style analysis can be supported using the Lag and Lead
functions provided with MicroStrategy. These functions can be used to
define metrics that compare values from different time periods without the
use of transformations. For information on creating transformations, see the
Project Design Guide.
l The examples shown below were created in the MicroStrategy Tutorial

project.
l The examples below use the Lag function. The Lead function can be used
in the same way. The difference in using Lead rather than Lag is that the
Lead function would display information by going forward in time rather
than backward. For example, while the Lag function can return historical
data from one year ago, the same formula used for the Lead function
would return historical data from one year ahead.
The Lead function does not predict future values. For functions that can
predict future values based on historical data, see ForecastV (Forecast,
Vector Input), page 453, GrowthV (Growth, Vector Input), page 459, and
TrendV (Trend, Vector Input), page 488. Additionally, you can use data
mining functions from MicroStrategy Data Mining Services to predict
future trends, as described in Data Mining Functions, page 343.
For example, a Last Year's Monthly Revenue metric can be created using
the following function:
Lag<SortBy=(Month@ID)>(Revenue,12,ZeroToNull(0))
This metric can then be included on a report with the Month attribute and the
Revenue metric, as shown below.

This allows you to perform a side-by-side comparison of monthly revenue for

different years. In this report, the 2007 monthly revenue is displayed next to
the 2008 monthly revenue. The element Jan 2009 is included to show that
the Last Year's Monthly Revenue is displaying the monthly revenue from the
previous year.
By modifying the offset of the Lag function, you can change the time periods
that can be compared side by side. For example, the function listed above
uses an offset of 12, which displays monthly data from one year ago. Using
an offset of 3 would display monthly data from three months ago, while using
an offset of 24 would display monthly data from two years ago. These slight
modifications could be used to create separate metrics that could all be
included on the same report. The report below shows an example of
including three different metrics that use the same Lag formula with a
different offset.

The report shown above allows side-by-side comparison of monthly revenue

for three different time periods. In this report, the Monthly Revenue 2 Years
Ago metric displays no data for months in the year 2008 because no revenue
data exists for the year 2006.
To use the Lag or Lead functions for transformation-style analysis, the

metric formulas must be created to support the required reporting scenario.
For example, the report with the Month attribute and the Revenue metric has
the Category attribute added to it. To support this reporting scenario, you
can modify the Lag formula described above to include the Category
attribute, as shown below.
Lag<BreakBy={Category},SortBy=(Month@ID)> (Revenue,12,ZeroToNull
(0))

Using the formula shown above, the calculation is restarted for each
category, which allows the side-by-side comparison of monthly revenue over
time for each category, as shown in the report below:
Only the Month attribute elements Jan 2007, Jan 2008, and Jan 2009 are
displayed to show that the Last Year's Monthly Revenue (Category) metric is
displaying the monthly revenue from the previous year.
LastInRange
Returns the last value in a range of values. LastInRange can be used to
examine data such as inventory at the end of each month.
Calculations can be restarted based on attributes set in the function

parameters. This restart capability differentiates LastInRange from Last.
While Last is a group-value function and takes both facts and metrics as
argument input, LastInRange is an OLAP function and takes only metrics
as argument input.
Syntax
LastInRange <BreakBy={}, SortBy=()> (Argument)
Where:

Expression
Where:
l k = row number
l y k = last non-NULL value of the metric after applying the BreakBy and
SortBy parameters
Lead
The Lead function is useful to display a set of metric values in an order
determined by another metric or attribute on the report. This allows you to
compare metric values side-by-side. The easiest way to understand the
Lead function is with an example, provided below.
Syntax
Lead <BreakBy, SortBy> (Argument, Offset, DefaultValue)
Where:
l Argument is an attribute or metric representing a list of values. It is

common to use a metric for Argument.
l Offset specifies the offset from the current row to display information for.
This offset precedes ahead of the current row (you can use the Lag
function to use an offset that trails behind the current row, see Lag, page

200). For example, with an offset of two, row one displays data for the row
two ahead of it, which would be row three.
l DefaultValue is the value displayed for any entries that do not have any
data due to the offset. For example, when using an offset of 2, the last two
entries use the default value.
Some common default values to display include:
l 0: To display a value of 0 for any entries that do not have any data due
to the offset, type 0.
l No data: To display no data for any entries that do not have any data due
to the offset, type ZeroToNull(0).
Example
For an example of using Lag and Lead functions to perform transformation-

style analysis, see Transformation-Style Analysis Using the Lag and Lead
Functions, page 203.
On a report with Item, Cost, and Profit, you can use the Lead function to
create a Cost (Lead on Profit) metric. This metric displays Cost values
based on the descending sort order of the Profit metric, and is defined with
the following expression:
Lead<SortBy= (Profit Desc) >(Cost, 3.0, 0.0)
Notice that the offset is set to three, meaning that the display of cost values
is displayed three ahead of the current value. This is displayed in the report
results shown below.
Only the top and the bottom of the report is shown. To view the entire report
results, you can create this report in the MicroStrategy Tutorial project.

The Cost (Lead on Profit) values are displayed three ahead of the Cost
metric, and are displayed based on the descending sort order of Profit.
Notice that with an offset of three, the last three entries for Cost (Lead on
Profit) use the default value of zero. Also, the first three values of Cost are
not included in the Cost (Lead on Profit) metric.
The report has been sorted based on the Profit metric to make the Cost
(Lead on Profit) values easier to analyze.
MovingAvg (Moving Average)

Returns the moving average of the current value and preceding values, as
defined by the WindowSize parameter. The calculations can be restarted
based on attributes set in the function parameters. This is an OLAP function.
Syntax
MovingAvg <BreakBy={}, SortBy=()> (Argument, WindowSize)

Where:
l WindowSize is a positive integer indicating the number of values to use in

each calculation.
Expression
Where:
l y i = metric value at the i th row
l m= window size
l n = number of rows/metric values
Example
This simple example illustrates how the MovingAvg function calculates a

list of values and returns the average of a specified number of values. In this
case, the window size is set to 3, meaning that the value in the
MovingAverage column represents the average of the current value of the
two values that precede it in the value list. The calculation is shown in the
following table.
Values MovingAverage
10 10 (10/1)

Values MovingAverage
20 15 ((20+10) / 2)
30 20 ((30+20+10) / 3)
15 51.67 ((15+30+20) / 3)
5 16.67 ((5+15+30) / 3)
20 13.34 ((20+5+15) / 3)
40 21.67 ((40+20+5) / 3)
As an additional example, the Human Resources Analysis Module project

includes the Division Breakdown report shown below.
This report displays details about employee headcounts for each division,
over the various quarters of 2010. Included in this report is the metric
MovingAvg. It is defined as a derived metric, using the following expression:
MovingAvg<BreakBy={Division}, SortBy=(Quarter)> (Employees, 4.0)

This expression calculates the moving average of employee headcount for a

given division, during the four quarters of 2010. The window size of 4.0
specifies that the average is calculated across the four quarters, and
BreakBy={Divison} ensures that the moving average calculation is
specific to each division.
Using this MovingAvg metric, you can determine that the Sales division had
between 13.0 and 16.3 employees on average during the 2010 year, with
slight increases throughout the year.
For an extension of this example on how you can also use a weighted
moving average, see ExpWghMovingAvg (Exponential Weighted Moving
Average), page 192.
MovingCount
Returns the moving count of a list of values. The calculation can be
restarted based on attributes set in the function parameters. This is an
OLAP function.
Syntax
MovingCount <BreakBy={}, SortBy=()> (Argument, WindowSize)
Where:
l Argument is a metric representing a list of values.
l WindowSize is a positive integer indicating the highest number to use in

the count.
Expression

Where:
l m= window size
l 1 i = 1 if the value at the i th row is not NULL
l 1 i = 0 otherwise
Usage Notes
If there are more entries in a section than the integer in the window size
parameter, then all the remaining entries are given the highest number in the
count. For example, if the window size is 4 and there are 6 entries in the list
section, they are counted as follows: 1,2,3,4,4,4.
Example
This simple example demonstrates how the MovingCount function counts

rows of data. This report uses the attributes Region and Employee, and the
metrics Revenue and Moving Count. A filter is applied so the only Regions
displayed are South, Northwest, and Southwest. The Moving Count metric is
defined as follows:
MovingCount<BreakBy={Region}, SortBy= (Value) >(Revenue, 3.0)
l The count restarts for every Region.

l The entries are counted based on the value of the metric Revenue in
ascending order (the lowest value is counted as 1, next lowest is 2, and so
on).
l The highest number in the count is 3 as designated in the WindowSize

parameter.
MovingDifference
Returns the moving difference between current and preceding values. The
position of the preceding row used in the calculation is defined in the
function arguments. The calculations can be restarted based on attributes
set in the function parameters. This is an OLAP function.
Syntax
MovingDifference <BreakBy, SortBy> (Argument, WindowSize)
Where:
l WindowSize is a positive integer indicating the range of values used to

compute each difference.
Expression
Where:
l m = window size
Example
This example illustrates how the MovingDifference function subtracts the

value of a preceding row from the value of the current row and returns the
difference. In this case, the window size is set to 3, meaning that the value
two rows previous to the current row is subtracted from the current row. In
other words, there are 3 rows in the window, the current row is 3, the row
immediately preceding it is 2, and the row preceding that is 1; therefore, the
calculation for the moving difference of row 3 is (row 3 - row 1).
The value in the MovingDifference column represents the result of the

calculation for every window of three values. The calculation is also shown.
Values MovingDifference
400
500
700 300 (700-400)
300 -200 (300-500)
600 -100 (600-700)
800 500 (800-300)
200 -400 (200-600)

MovingMax (Moving Maximum)

Returns the moving maximum value by comparing current and preceding
rows as defined in the function arguments. The calculation can be restarted
based on attributes set in the function parameters. This is an OLAP function.
Syntax
MovingMax <BreakBy, SortBy> (Argument, WindowSize)
Where:
l WindowSize is a positive integer indicating the number of values to

compare in each calculation.
Expression
Where:
l m = window size
Example
This simple example illustrates how the MovingMaximum function moves

through a list of values, subtracting a value from the user-defined number of
preceding rows. In this case, the window size is set to 3, meaning that the
value in the MovingMaximum column represents the highest value among

the current and preceding two values in the value list. The calculation is
shown in the following table.
Values MovingMaximum
550 550
30 550 (550>30)
40 550 (550>40 and 550>30)
70 70 (70>40 and 70>30)
50 70 (70>50 and 70>40)
MovingMin (Moving Minimum)

Compares the current value and preceding values as defined in the function
arguments to calculate a moving minimum value. The calculation can be
OLAP function.
Syntax
MovingMin <BreakBy,SortBy>(Argument, WindowSize)
Where:

each calculation.
Expression

Where:
l m = window size
Example
This simple example illustrates how the MovingMin function moves through
a list of values returning the lowest value within a specified number of
values. In this case, the window size is set to 3, meaning that the value in
the MovingMinimum column represents the lowest value among the current
and preceding two values in the value list. The calculation is shown in the
following table.
Values MovingMinimum
550 550
30 30 (30<550)
40 30 (30<40 and 30<550)
70 30 (30<70 and 30<40)
50 40 (40<70 and 40<50)
MovingStDevP (Moving Standard Deviation of a Population)

Returns the moving standard deviation of a population based on a list of
values that encompasses the whole population. The calculation can be
OLAP function.
Syntax
MovingStDevP <BreakBy, SortBy> (Argument, WindowSize)
Where:


each calculation.
Expression
Where:
l y = average of metric
l m= window size
Example
This example shows a report where the moving standard deviation of the
revenue is calculated. This calculation is based on the assumption that the
list of values supplied in the metric represents the entire population of the
data for which you want to obtain the standard deviation. The calculation
starts over for each region, the information is sorted within the region by
state in ascending order, and each calculation is based upon a window size
of 3.
Compare this example report to the example for MovingStDev to see the
different values returned when calculating for a population as opposed to a
sample.

and the metrics Total Revenue, MovingStDevP, RunningStDevP, and
StDevP. A report filter limits data to the South, Southwest, and Northwest
regions. The definition of the MovingStDevP metric is as follows:
MovingStDevP<BreakBy={[Customer Region]}, SortBy=<[Customer

State])>([Total Revenue],3)
MovingStDev (Moving Standard Deviation)

Returns the moving standard deviation based on a list of values that is a
sample of the population. The calculation can be restarted based on
attributes set in the function parameters. This is an OLAP function.
Syntax
MovingStDev <BreakBy, SortBy> (Argument, WindowSize)
Where:
l Argument is a metric representing a list of number.


each calculation.
Expression
Where:
l yi = metric value at the i th row
l y = average of metric
l m = window size
Example
This example shows a report where the moving standard deviation of the
list of values supplied in the metric represents a sample of the data for which
you want to obtain the standard deviation. The calculation starts over for
each region, the information is sorted within the region by state in ascending
order, and each calculation is based upon a window size of 3.
Compare this example report to the example for MovingStDevP to see the
sample.
and the metrics Total Revenue, MovingStDev, RunningStDev, and StDev. A
report filter limits data to the Southwest, Southeast, and Northwest regions.
The definition of the MovingStDev metric is as follows:

MovingStDev<BreakBy={[Customer Region]}, SortBy=<[Customer

State])>([Total Revenue], 3)
MovingSum
Returns the moving sum of the current value and preceding values as
defined in the function arguments. The calculations can be restarted based
on attributes set in the function parameters. This is an OLAP function.
Syntax
MovingSum <BreakBy,SortBy>(Argument, WindowSize)
Where:
l WindowSize is a positive integer indicating the number of values to sum

in each calculation.
Expression

Where:
l m = window size
Example
This simple example illustrates how the MovingSum function moves through
a list of values calculating and returning the sum of a specified number of
values. In this case the window size is set to 2 meaning that the sum in the
MovingSum column represents the current value added to the value from the
value list that precedes it. The calculation is shown in the following table.
Values MovingSum
10 10
20 30 (20+10)
30 50 (30+20)
15 45 (15+30)
5 20 (5+15)
OLAPAvg (OLAP Average)

Returns the average of the current value, preceding values, and succeeding
values as defined in the function arguments. Unlike RunningAvg and
MovingAvg, which can only include values above the current row in the

calculation, you can use OLAPAvg to include values both above and below
the current row in the calculation.
The calculations can be restarted based on attributes defined in the function

parameters.
Syntax
OLAPAvg<Distinct, OLAPWinStType, OLAPWinStOffset, OLAPWinEndType,

OLAPWinEndOffset, BreakBy, SortBy>(Argument)
Where:

the calculation or to calculate using only the unique values. If you define
Distinct to be true, then the parameters OLAPWinStType,
OLAPWinStOffset, OLAPWinEndType, OLAPWinEndOffset, and
SortBy are ignored.
l OLAPWinStType defines the window type for the starting location of the
calculation. Select one of the following options, listed by the name of the
setting and its corresponding OLAPWinStType value in parentheses:
l Top of data set (0): The calculation starts at the top value as determined
by the BreakBy and SortBy values.
l Current row (2): The calculation starts at the current row.
l N rows before current row (3): The calculation starts a number of rows
before the current row. You define this offset of rows with the
OLAPWinStOffset parameter described below.
l N rows after current row (4): The calculation starts a number of rows
after the current row. You define this offset of rows with the

l OLAPWinStOffset defines the offset of rows from the current row to start
the calculation. You can define this offset if the OLAPWinStType
parameter is defined as N rows before current row (3) or N rows after
current row (4).
l OLAPWinEndType defines the window type for the ending location of the
setting and its corresponding OLAPWinEndType value in parentheses:
l Bottom of data set (1): The calculation stops at the bottom value as
determined by the BreakBy and SortBy values.
l Current row (2): The calculation stops at the current row.
l N rows before current row (3): The calculation stops a number of rows
OLAPWinEndOffset parameter described below.
l N rows after current row (4): The calculation stops a number of rows
l OLAPWinEndOffset defines the offset of rows from the current row to

stop the calculation. You can define this offset if the OLAPWinEndType
current row (4).
The function is not valid if your starting point is at a lower point than your
ending point.
Example
An OLAPAvg Unit Cost metric is created using the OLAPAvg function based
on the Unit Cost metric, as defined below:

OLAPAvg<OLAPWinStType=3, OLAPWinStOffset=3, OLAPWinEndType=4,

OLAPWinEndOffset=2, BreakBy={Category} SortBy={Item}>([Unit
Cost])
The starting point for the average is defined as three rows before the current
row (OLAPWinStType=3, OLAPWinStOffset=3). The stopping point for
the average is defined as two rows after the current row
(OLAPWinEndType=4, OLAPWinEndOffset=2).
This metric is displayed on a report along with Category, Item, and Unit
Cost, as shown below.
There are a few facts about this data to take note of.
The first value of OLAPAvg Unit Cost is $15. This is calculated by adding
$32, $8, and $5 to get a total of $45. This total of $45 is then divided by
three (the number of rows included in the calculation) to get the final result
of $15. These rows are included because the calculation ends two rows after
the current row. Even though the calculation starts three rows before the
current row, there is no data above the current row to include in the
calculation.

The fourth value for OLAPAvg Unit Cost is the first value that can include
data from all three rows above the current row to two rows below the current
row in the calculation (($32 + $8 + $5 + $25 + $19 + $20) / 6 =
$18.17). Notice that with this calculation a value of six is used to calculate
the average because six rows were included in the average.
While it is not shown on this report, the calculation would restart for the first
Item of the next Category because the function is defined to break by the
Category attribute.
OLAPCount
Returns the count of the current value, preceding values, and succeeding
values as defined in the function arguments. Unlike RunningCount and
MovingCount, which can only include values above the current row in the
calculation, you can use OLAPCount to include values both above and below
the current row in the calculation.

parameters.
Syntax
OLAPCount<Distinct, Null, OLAPWinStType, OLAPWinStOffset,

OLAPWinEndType, OLAPWinEndOffset, BreakBy, SortBy>(Argument)
Where:

Distinct to be true, then the parameters Null, OLAPWinStType,
SortBy are ignored.
l Null is a TRUE/FALSE parameter that defines whether null values are

included in the total count.

current row (4).
l Bottom of data set (1): The calculation stops at the bottom value. The
top value is determined by the BreakBy and SortBy values.


current row (4).
ending point.
Example
An OLAPCount Unit Cost metric is created using the OLAPCount function

based on the Unit Cost metric, as defined below:
OLAPCount<OLAPWinStType=3, OLAPWinStOffset=3, OLAPWinEndType=4,

OLAPWinEndOffset=2, BreakBy={Category} SortBy={Subcategory}>
([Unit Cost])
The starting point for the count is defined as three rows before the current
row (OLAPWinStType=3, OLAPWinStOffset=3). The stopping point for
the count is defined as two rows after the current row
This metric is displayed on a report along with Category, Subcategory, and

Unit Cost, as shown below.

The first value of OLAPCount Unit Cost is three. This is calculated by

counting the current row and two rows after the current row. Even though the
calculation starts three rows before the current row, there is no data above
the current row to include in the calculation.
The fourth entry for OLAPCount Unit Cost is the first entry that can count all
three rows above the current row to two rows below the current row in the
calculation. This entry is able to reach the maximum value of six. Notice that
after this entry the count begins to decrease because there are no longer
two rows below the current row to include in the calculation.
The calculation restarts for the first Subcategory of the next Category
because the function is defined to break by the Category attribute.

OLAPMax (OLAP Maximum)

Returns the maximum value based on comparing the current value,
preceding values, and succeeding values as defined in the function
arguments. Unlike RunningMax and MovingMax, which can only include
values above the current row in the calculation, you can use OLAPMax to
include values both above and below the current row in the calculation.

parameters.
Syntax
OLAPMax<OLAPWinStType, OLAPWinStOffset, OLAPWinEndType,

Where:

current row (4).

current row (4).
ending point.
Example
An OLAPMax Unit Cost metric is created using the OLAPMax function based

OLAPMax<OLAPWinStType=3, OLAPWinStOffset=3, OLAPWinEndType=4,

([Unit Cost])
The starting point for the calculation is defined as three rows before the
current row (OLAPWinStType=3, OLAPWinStOffset=3). The stopping
point for the calculation is defined as two rows after the current row

The first value of OLAPMax Unit Cost is $14. This is calculated by returning
the maximum value of $14, $11, and $6. These are included because the

calculation ends two rows after the current row. Even though the calculation
starts three rows before the current row, there is no data above the current
row to include in the calculation.
The second value also returns $14 as it returns the maximum value of $14,
$11, $6, and $7.
The third value for OLAPMax Unit Cost is the first entry that includes $26 in
its maximum calculation. The rest of the OLAPMax Unit Cost values for the
Books Category return $26 because this is the maximum value, and it is
within the window of three rows above the current row to two rows below the
current row.
OLAPMin (OLAP Minimum)

Returns the minimum value based on comparing the current value,
preceding values, and succeeding values as defined in the function
arguments. Unlike RunningMin and MovingMin, which can only include
values above the current row in the calculation, you can use OLAPMin to
include values both above and below the current row in the calculation.

parameters.
Syntax
OLAPMin<OLAPWinStType, OLAPWinStOffset, OLAPWinEndType,

Where:

current row (4).


current row (4).
ending point.
Example
An OLAPMin Unit Cost metric is created using the OLAPMin function based
OLAPMin<OLAPWinStType=3, OLAPWinStOffset=3, OLAPWinEndType=4,

([Unit Cost])
The starting point for the calculation is defined as three rows before the
point for the calculation is defined as two rows after the current row


The first value of OLAPMinUnit Cost is $6. This is calculated by returning the
minimum value of $14, $11, and $6. These are included because the
calculation ends two rows after the current row. Even though the calculation
starts three rows before the current row, there is no data above the current
row to include in the calculation.
All of the OLAPMin Unit Cost values for the Books Category return $6
because this is the minimum value, and it is within the window of three rows
above the current row to two rows below the current row.

OLAPRank
Returns the rank of the current value based on the other values defined by
the sorting criteria. The ranking can be restarted based on attributes defined
in the function parameters.
The OLAPRank function allows the ranking to be calculated in the database,

rather than calculating the ranking using the MicroStrategy Analytical
Engine.
l Since OLAPRank is calculated in the database, you can only use this
function if your database supports the OLAPRank function.
l Any metrics that use the OLAPRank function must not be defined as smart
metrics. If metrics that use OLAPRank are defined as smart metrics, the
calculation is performed in the MicroStrategy Analytical Engine and an
error is returned.
l You must include at least one metric in the SortBy parameter.
Syntax
OLAPRank<BreakBy, SortBy>()
Where:
l BreakBy is the attribute indicating where the ranking restarts.
l SortBy is the attribute or metric by which the data is sorted. For

OLAPRank, the SortBy parameter is also where you can include the
metric to perform the calculation on. You must include at least one metric
in the SortBy parameter.
Example
The MicroStrategy Tutorial project includes an Avg Revenue per Customer

metric. This metric can be placed on a report along with the attributes
Quarter and Region to display the quarterly average revenue per customer
for each region.

To extend this analysis, you can create an OLAPRank Avg Rev per
Customer metric as defined below:
OLAPRank<BreakBy={Quarter}, SortBy= ([Avg Revenue per Customer],

Region@ID) >()
When this metric is included on a report with Quarter, Region, and Avg
Revenue per Customer, it displays the regional rank of the quarterly average
revenue per customer for each region. This is shown in the report below.
The report has been sorted by Quarter, and then by the OLAPRank Avg Rev
per Customer values.
The report shown above displays regions based on their average revenue
per customer for each quarter, sorted from the lowest average revenue per

customer to the highest. This analysis allows you to quickly see how regions
are performing based on average revenue per customer over different
quarters. In the first three quarters of data shown above, Web has the lowest
average revenue per customer. However, there is some variation in the
performance of the other regions. Performing this analysis over extended
periods of time can help to show trends in revenue and regional
performance.
OLAPSum
Returns the summation of the current value, preceding values, and
succeeding values as defined in the function arguments. Unlike
RunningSum and MovingSum, which can only include values above the
current row in the calculation, you can use OLAPSum to include values both
above and below the current row in the calculation.

parameters.
Syntax
OLAPSum<Distinct, OLAPWinStType, OLAPWinStOffset, OLAPWinEndType,

Where:

Distinct to be true, then the parameters OLAPWinStType,
SortBy are ignored.

current row (4).
l N rows before current row (3): The calculation stops a number of rows
l N rows after current row (4): The calculation stops a number of rows



current row (4).
ending point.
Example
An OLAPSum Unit Cost metric is created using the OLAPSum function

based on the Unit Cost metric, as defined below:
OLAPSum<OLAPWinStType=3, OLAPWinStOffset=3, OLAPWinEndType=4,

([Unit Cost])
The starting point for the summation is defined as three rows before the
point for the summation is defined as two rows after the current row


The first value of OLAPSum Unit Cost is $30.82. This is calculated by adding
$13.93, $10.75, and $6.13. These are included because the calculation ends
two rows after the current row. Even though the calculation starts three rows
before the current row, there is no data above the current row to include in
the calculation.
The data displayed on the report is rounded to the nearest cent, which can
give the impression that some calculations are slightly incorrect. For
example, adding $13.93, $10.75, and $6.13 actually totals $30.81 rather
than $30.82 as displayed on the report. This difference is because the data
is rounded up for display on the report. You can display more decimal
values for the Unit Cost and OLAPSum Unit Cost metrics to see the exact
values.

The fourth value for OLAPSum Unit Cost is the only value that can include
data from all three rows above the current row to two rows below the current
row in the calculation ($13.93 + $10.75 + $6.13 + $7.20 + $25.93
+ $9.43 = $73.38).
The final OLAPSum Unit Cost value for the Books Category can only include
the Unit Cost value for the current row and the three rows above it ($6.13 +
$7.20 + $25.93 + $9.43 = $48.70). It cannot include two rows below
the current row because the calculation restarts for the first Subcategory of
the next Category. The calculation restarts because the function is defined
to break by the Category attribute.
RunningAvg (Running Average)

Moves through the values in a list and returns the running average, adding
the current value to the sum of the preceding values and dividing by the
current count of values. The calculation can restart based on attribute
groupings identified in the parameter settings. This is an OLAP function.
Syntax
RunningAvg <BreakBy,SortBy>(Argument)
Where:
Expression
Where:

l m = window size
Example
This simple example illustrates how the RunningAverage function walks

through a list of values calculating and returning the new average with the
addition of each value. The calculation is shown in the following table.
Values RunningAverage
10 10 (10/1)
20 15 (30/2)
30 20 (60/3)
40 25 (100/4)
50 30 (150/5)
RunningCount
Returns the running count for each value in a list of values, returning the
current count after each value. The count can be restarted based upon
attributes identified in the parameter settings. This is an OLAP function.
Syntax
RunningCount<BreakBy,SortBy>(Argument)
Where:
l Argument is a metric representing the list of values.
Expression

Where:
l 1 i = 0 if the i th row of argument is NULL
1 i = 1 otherwise
Example
This simple example demonstrates how the RunningCount function counts

rows of data. This report uses the attributes Region and Employee, and the
metrics Revenue and Running count. A filter is applied so the only Regions
displayed are South, Northwest, and Southwest. The Running count metric
is defined as follows:
RunningCount<BreakBy={Region}, SortBy= (Value) >(Revenue)
l The count restarts for every Region.
l The entries are counted based on the value of the metric Revenue in
ascending order (the lowest value is counted as 1, next lowest is 2, and so
on).

RunningMax (Running Maximum)

Returns the running maximum value in a list of values by comparing the
current and preceding values. The evaluation can restart based on attributes
identified in the parameter settings. This is an OLAP function.
Syntax
RunningMax <BreakBy, SortBy>(Argument)
Where:
l Argument is a metric representing the list of numbers.
Expression
Where:
Example
This simple example illustrates how the RunningMax function moves through
a list of values comparing each value to the highest value identified so far
and returning the maximum value as it progresses. The calculation is shown
in the following table.
Values RunningMaximum

Values RunningMaximum
6 8 (8>6)
10 10 (10>8)
9 10 (10>9)
5 10 (10>5)
RunningMin (Running Minimum)

Returns the running minimum value in a list of values by comparing current
and preceding values. The evaluation can restart based on attributes
identified in the parameter settings. This is an OLAP function.
Syntax
RunningMin <BreakBy, SortBy> (Argument)
Where:
Expression
Where:
Example

This simple example illustrates how the RunningMinimum function walks

through a list of values comparing each value to the lowest value identified
so far and returning the minimum value as it progresses. The calculation is
shown in the following table.
Values RunningMinimum
6 6 (6<8)
10 6 (6<10)
9 6 (6<9)
5 5 (5<6)
RunningStDevP (Running Standard Deviation of a Population)

Returns the running standard deviation of a population for a value
expression. The list of values supplied is the population. The calculation can
restart based on attributes identified in the function parameter settings. This
is an OLAP function.
Syntax
RunningStDevP <BreakBy, SortBy> (Argument)
Where:
Expression

Where:
Example
This example shows a report where the running standard deviation of the
list of values supplied in the metric represents the entire population of the
data for which you want to obtain the standard deviation. The calculation
starts over for each region, and the information is sorted within the region by
state in ascending order.
Compare this example report to the example for RunningStDev to see the
sample.
and the metrics Total Revenue, RunningStDevP, and StDevP. A report filter
limits data to the Southwest, Southeast, and Northwest regions. The
definition of the RunningStDevP metric is as follows:
RunningStDevP<BreakBy={[Customer Region]}, SortBy=<[Customer

State])>([Total Revenue])

RunningStDev (Running Standard Deviation)

Returns the running standard deviation of a sample for a value expression.
The list of values supplied is the sample. The calculation can restart based
on attributes identified in the function parameters. This is an OLAP function.
Syntax
RunningStDev<BreakBy,SortBy>(Argument)
Where:
Expression
Example

This example shows a report where the running standard deviation of the
list of values supplied in the metric represents a sample of the data for which
you want to obtain the standard deviation. The calculation starts over for
each region, and the information is sorted within the region by state in
ascending order.
Compare this example report to the example for RunningStDevP to see the
sample.
and the metrics Total Revenue, RunningStDev, and StDev. A report filter is
used to limit the data to the Southwest, Southeast and Northwest regions.
The definition of the RunningStDev metric is as follows:
RunningStDev<BreakBy={[Customer Region]}, SortBy=<[Customer

RunningSum
This function returns the running sum of the values in a list, adding the
current value to the preceding values. It can be used to maintain running

totals of values in specific categories as well as an entire set. The

calculation can restart based on attributes identified in the parameter
settings. This is an OLAP function.
Syntax
RunningSum <BreakBy,SortBy>(Argument)
Where:
l BreakBy is the parameter that sets the attribute designating where the
calculation restarts.
l SortBy is the parameter that sets the attribute or metric by which the data
is sorted.
Expression
Example
Example 1: A running sum displays the results of each calculation as it

works towards the total of a set of values. Using the value set (1, 2, 3, 4, 5),
the following table illustrates the running sum and its calculations.
Values RunningSum
1 1 (0+1)
2 3 (1+2)
3 6 (3+3)
4 10 (6+4)
5 15 (10+5)

Example 2: This example shows a report where the running sum of the
revenue is calculated. The calculation starts over for each region, and the
information is sorted within the region by state in ascending order. The
report contains the attributes Customer Region and Customer State, and the
metrics Total Revenue and Running Sum. A report filter limits data to the
Southwest, Southeast, and Northwest regions. The definition of the Running
Sum metric is as follows:
RunningSum<BreakBy={[Customer Region]}, SortBy=<[Customer

Note that the subtotaled Total Revenue is equal to the last value in the
Running Sum column for each region.
WeightedCorr (Weighted Correlation)

A correlation of two values describes the degree to which the values are
related or associated. Values that are closely related with either a positive

or negative correlation will have a correlation close to 1 or -1 respectively,

while values that are not correlated at all will have a correlation close to 0.
A weighted correlation allows you to apply a weight, or relative significance

to each value comparison. Correlation comparisons with a higher value for
their weight are considered as more significant when compared to the other
value comparisons.
To determine the correlation of two values without applying a weight to the

comparisons, see Correlation, page 446.
Syntax
WeightedCorr <FactID> (Argument1, Argument2, Weight)
Where:

l Argument1 and Argument2 are attributes, facts, or metrics representing

lists of numbers for comparison. A correlation is calculated on these
values to determine the level of association between the two values.
l Weight is an attribute, fact, or metric representing a list of numbers to

define the weight of each comparison.
Expression
Where:
l x and y are the values being compared.
l w is the weight applied to each comparison.
Usage Notes

l The correlation coefficient is measured on a scale that varies from 1 to - 1.

So only a value between -1 and 1 is returned.
l Complete correlation between two variables is expressed by either 1 or -1.

When one variable increases as the other increases, the correlation is
positive. When one decreases as the other increases, the correlation is
negative. Complete absence of correlation is represented by 0.
l If an array or reference argument contains text, logical values, or empty

cells, those values are ignored; however, cells with the value zero are
included.
l If Argument1 and Argument2 have a different number of data points, an

error is returned.
l If either Argument1 or Argument2 is empty, or if the standard deviation

of their values equals zero, an error is returned.
Example
Your company keeps employee statistics including their overall satisfaction

with their job and role at the company, as well as their performance score.
While these are both important statistics, it can also be beneficial to know
how closely related these two statistics are. Knowing this information can
help determine if an employee's satisfaction with their job is related to their
performance.
A report including both the correlation and weighted correlation of these

statistics, displayed as percentages, is shown below for employees at the
executive level:

The expressions for these calculations are as follows:
l Correlation:
Correlation([Avg. Performance Score], [Avg. Satisfaction

Score]) {Level , ~ }
l Weighted Correlation:
WeightedCorr([Avg. Performance Score], [Avg.

Satisfaction Score], Tenure) {Level, ~}
Both the correlation and weighted correlation indicate that employee

satisfaction and performance are positively correlated, meaning that when
one increases so does the other. The weighted correlation includes an
employee's tenure into the correlation calculation. This means that more
significance is given to correlation comparisons for employees that have
been with the company longer.
For the executive level employees, factoring in tenure results in a smaller

correlation between satisfaction and performance. This type of analysis can
be crucial in determining how performance can be improved or maintained
both for new hires and long tenured employees.
WeightedCov (Weighted Covariance)

Covariance is used to examine the relationship between two data sets. For
instance, the covariance can be used to examine whether an increase in
income is related to higher education levels. A covariance greater or less
than zero indicates a relationship, while a value of zero indicates no
relationship.
A weighted covariance allows you to apply a weight, or relative significance

to each value comparison. Covariance comparisons with a higher value for
value comparisons.

To determine the covariance of two values without applying a weight to the

comparisons, see Covariance, page 447.
Syntax
WeightedCov <FactID> (Argument1, Argument2, Weight)
Where:
l Argument1 and Argument2 are attributes, facts, or metrics representing

lists of numbers for comparison. A covariance is calculated on these
values to determine the level of association between the two values.

define the weight of each comparison.

Expression
l x and y are the Argument1 and Argument2 values being compared.
l w is the weight applied to each comparison.
l
= average value of x
= average value of y
l
Usage Notes
The following are invalid conditions:
l If an array or reference argument contains text, logical values, or empty

included.

l If Argument1 and Argument2 have a different number of data points, an

error is returned.
l If either Argument1 or Argument2 is empty, an error is returned.
Example
Your company keeps employee statistics including their overall satisfaction

with their job and role at the company, as well as their performance score.
While these are both important statistics, it can also be beneficial to know
how closely related these two statistics are. Knowing this information can
help determine if an employee's satisfaction with their job is related to their
performance.
A report including both the covariance and weighted covariance of these

statistics, is shown below for employees at the executive level:
The expressions for these calculations are as follows:
l Covariance:
Covariance([Avg. Performance Score], [Avg. Satisfaction

Score]) {Level , ~ }
l Weighted Covariance:
WeightedCov([Avg. Performance Score], [Avg. Satisfaction

Score], Tenure) {Level, ~}
Both the covariance and weighted covariance indicate that employee

satisfaction and performance are positively related, meaning that when one

increases so does the other. The weighted covariance includes an

employee's tenure into the covariance calculation. This means that more
significance is given to covariance comparisons for employees that have
been with the company longer.
For the executive level employees, factoring in tenure results in a smaller

relationship between satisfaction and performance. This type of analysis can
be crucial in determining how performance can be improved or maintained
both for new hires and long tenured employees.
WeightedMean
An average, also known as an arithmetic mean, is the sum of a set of values
divided by the number of values in the set.
A weighted mean allows you to apply a weight, or relative significance to

each value when determining an average. Values with a higher value for
values.
To calculate an average without applying a weight to the values, see Avg

(Average), page 117.
Syntax
WeightedMean<FactID>(Argument1, Weight)
Where:
l Argument1 is an attribute, fact, or metric representing a list of numbers

that are used to calculate an average.

define the weight of each value.

Expression

Example
You can calculate a weighted mean to apply greater significance to certain

values used to determine an average. For example, you can use the percent
growth of revenue as a way to define revenue values as more or less
significant for an average. The report shown below contains the attribute
Customer Region and the metrics Revenue, Percent Growth, Average
Revenue, and Weighted Mean Revenue. The report also includes the Year
attribute in the page-by to show results by year.
The expressions for the average and weighted mean calculations are as
follows:
l Average Revenue:
Avg(Revenue){Year}
l Weighted Mean Revenue:
WeightedMean(Revenue, [Percent Growth]){Year}
The weighted mean takes into account the weights of each revenue value.
For example, since Central's percent growth is greater than Northwest's
percent growth, and the weight in this example is the percent growth,
Central's revenue is given more significance in the weighted mean
calculation. This results in $2,229,195 for the Weighted Mean Revenue as
compared to $2,122,695 for the Average Revenue for all customer regions
for the given year.

Notice that while the Average Revenue and Weighted Mean Revenue are
applicable to a year in this example, the values are displayed for every
Customer Region row. One way to simplify the display of this information is
to include this data on a dossier. You can include the Average Revenue and
Weighted Mean Revenue results in text fields along with additional
visualizations of the data, as shown below.
WeightedStDev (Weighted Standard Deviation of a Sample)

The standard deviation is an indicator of how widely values in a group differ
from the mean (see StDev (Standard Deviation of a Sample), page 142). It is
useful for comparing different sets of values with a similar mean.
A weighted standard deviation allows you to apply a weight, or relative

significance to each value in a set of values. Values with a higher value for
their weight are considered as more significant to a sample as compared to
the other values in a sample.
WeightedStDev returns the weighted standard deviation of a population

based on a sample. This is a group-value function.
Syntax
WeightedStDev <Distinct, FactID>(Argument, Weight)
Where:

l Weight is an attribute, fact, or metric representing a list of numbers to define the

weight of each value.

Expression
Where:
l w i : The weight of the ith value. Values with a higher value for their weight
are considered as more significant to a sample as compared to the other
values in a sample.
l N': The number of weights that are not equal to zero.
l x w : The weighted mean of the values.
Usage Notes
l In this function, arguments correspond to a population sample as opposed

to the entire population. For entire populations, see StDevP (Standard
Deviation of a Population), page 140.
Example
This example shows a report where the standard deviation and a weighted
standard deviation of the revenue are calculated. This calculation is based
on the assumption that the list of values supplied in the metric represents a
sample of the data for which you want to obtain the standard deviation. The
calculation is based on the revenue values for each state within a region and
calculated at the region level.

and the metrics Revenue, Standard Deviation, and Weighted Standard
Deviation.
The definition of the standard deviation metrics are as follows:
l Standard Deviation:
StDev(Revenue){[Customer Region], ~}
l Weighted Standard Deviation:
WeightedStDev(Revenue, Revenue) {[Customer Region] , ~ }
For this example, the revenue values are also used as the weights given
to each revenue value included in the standard deviation.
The report is shown below:
The weighted standard deviation also takes into account the weights of each
revenue value. For example, since Connecticut's revenue is greater than
Maine's revenue, and the weight in this example is the revenue value,
Connecticut's revenue is given more significance in the weighted standard

deviation calculation. This results in $957,689 for the standard deviation as

compared to $1,141,237 for the weighted standard deviation of Northeast
revenue.
Notice that while the standard deviation and weighted standard deviation are
applicable to a Customer Region in this example, the values are displayed
for every Customer State row. One way to simplify the display of this
information is to include this data on a dossier. You can include the standard
deviation and weighted standard deviation results in text fields along with
additional visualizations of the data, as shown below.
Pattern Functions
Sequential Pattern Matching (SPM) takes an input table, sorts it in a
specified sequence, then scans through the rows looking for a pattern
across rows, and then returns aggregate information based on the matching
rows.
PatternMatch
This function defines a transaction in terms of a Derived Attribute. One or
more attribute elements or metric values will be mapped to the same
transaction group (assign the same integer number) if matching with a given
pattern.

Syn t ax
PatternMatch<Breakby, SortBy>(ValueList, Pattern, maxNum)
Where:
ValueList is either be a list of attribute elements or metric values.
Pattern is a pattern for given input to be matched with. Such pattern is

defined using a regular expression or simply {%string} to match with any
string.
maxNum is the maximum number of elements allowed within the same

transaction group (assigned with the same integer number). "-1" value
indicates no limitation.
PatternMatchCorr
This function defines a transaction in terms of a Derived Attribute. Given
attribute elements or metric values are mapped to the same transaction
group only if input1 (ValueList1) matches with a pre-defined pattern and
input2 (ValueList2) qualifies on certain conditions.
Syn t ax
PatternMatchCorr<Breakby, SortBy>(ValueList1, Pattern,

ValueList2, maxPairDiff, maxDiff, maxNum)
Where:
l ValueList1/ValueList2 are either lists of attribute elements or metric

values.
l Pattern specifies a pattern for ValueList1 to be matched with. Such

pattern is defined using a regular expression or simply {%string} to match
with any string.
l maxPairDiff specifies the maximum difference allowed between the

neighbor elements within ValueList2 to be mapped to the same

transaction group (assign the same integer number). "-1" value indicates
no limitation.
l maxDiff specifies the maximum difference between the minimum and the
maximum elements within ValueList2 to be mapped to the same
transaction group (assign the same integer number). "-1" value indicates
no limitation.
l maxNum specifies the maximum number of elements allowed within the

same transaction group (assigned with the same integer number). "-1"
value indicates no limitation.
PatternMatchDiff
This function defines a transaction in terms of a Derived Attribute. One or
more given attribute elements or metric values are mapped to the same
transaction group (assign the same integer number) if certain conditions
(specified by parameters: maxPairDiff, maxDiff and maxNum) are met.
Syn t ax
PatternMatchDiff<Breakby, SortBy>(ValueList, maxPairDiff,

maxDiff, maxNum)
Where:
l ValueList can either be a list of attribute elements or metric values.
l maxPairDiff specifies the maximum difference allowed between the

neighbor elements within the ValueList to be mapped to the same
transaction group (assigned with the same transaction number). "-1" value
indicates no limitation.
l maxDiff specifies the maximum difference between the first and the last
elements within the ValueList to be mapped to the same transaction group
(assigned with the same transaction number). "-1" value indicates no
limitation.

l maxNum specifies the maximum number of elements allowed within the

same transaction group (assigned with the same integer number). "-1"
value indicates no limitation.
Rank and NTile Functions

Rank and NTile functions are used to qualify a list of values relative to the
other values. For example, out of the four quarters in a year, you want to
rank total revenue from one to four, with one being the top revenue for the
year.
Functions including Rank, Percentile, PercentRank, and

PercentRankRelative allow you to view the ranking of values such as a
simple integer list or various types of percentages.
NTile functions are used to group the values in an ordered list into one of
several buckets or NTiles. Each element in the list is assigned an integer
corresponding to the bucket to which it belongs. The various NTile functions
differ in how the buckets are defined. Some functions allow you to define the
number of buckets, others allow you to define the size of the buckets, and so
on.
NTile
NTile functions are used to group the values in an ordered list into one of
several buckets or NTiles. For the NTile function, the buckets are
calculated so that each bucket has exactly the same number of rows
assigned to it or at most one row more than the others (the exception is that
identical value rows are placed in the same bucket). For example, if you
have 100 rows in a partition and define an NTile function with four buckets,
25 rows will be assigned a value of 1, 25 rows will have value 2, and so on.
These buckets are referred to as equiheight buckets.
If the number of rows in the partition does not divide evenly into the number
of buckets, then, barring identical value rows, the number of rows assigned

per bucket will differ by one at most. The extra rows are added to buckets
using the calculations ceiling(1*(buckets/remainder)),...,
ceiling(remainder*(buckets/remainder)).
For example, if there are 103 distinct value rows in a partition which has an
NTile<Tiles=5>() function, the first 20 rows will be in the first bucket, the
next 21 in the second bucket, the next 20 in the third bucket, the next 21 in
the fourth bucket, and the final 21 in the fifth bucket. The calculations
ceiling(1*(5/3))=2, ceiling(2*(5/3))=4, and ceiling(3*
(5/3))=5 include one of the extra three rows each in the second, fourth,
and fifth buckets.
NTile distributes the values in the specified metric, sorted in either

ascending or descending order, over a user-defined number of buckets.
Each bucket has an equal number of elements (if possible). This is an OLAP
function.
Syntax
NTile <Ascending, Tiles, BreakBy> (Argument)
Where:
l Ascending is a TRUE/FALSE parameter that designates the organization

of data within the NTiles.
l BreakBy is the parameter that sets the attribute determining where the
l Tiles is a positive integer that designates the number of buckets or

NTiles.
l Argument is a metric representing a list of values to be distributed equally

(if possible) into n buckets.
Example
Example 1: If you define Tiles=4 for a metric that contains 20 values, the
function distributes the numbers as follows:

l Values 1 through 5 in bucket 1
Example 2: This example shows a report where the customer states were
sorted based on revenue and then divided among a specified number of
buckets. The number of buckets is defined as 8. The report includes the
attribute Customer State and the metrics Total Revenue and NTile. The
NTile metric is a derived one, the syntax for which is as follows:
NTile<Tiles=8>([Total Revenue])
The resulting report divides the 48 states into 8 NTiles, each containing 6
elements (states). Within each band the data is sorted in ascending order by
the attribute Customer State. A portion of the report is displayed as follows.

NTileSize
NTileSize distributes the values in the specified metric, sorted in either
ascending or descending order, with the same number of elements in each
bucket. The number of elements in each bucket is user-defined. This is an
OLAP function.
Syntax
NTileSize <Ascending, BreakBy> (Argument, Size)
Where:


l Argument is a metric representing a list of values to be distributed in

buckets.
l Size is a positive integer that designates the number of elements per

bucket.
Example
Example 1: If you define the Size as 2, the function returns buckets

containing two values each. If there are six values, values one and two go in
bucket 1, values three and four go in bucket 2, and values five and six go in
bucket 3.
Example 2: This example shows a report where the customer states are
sorted by revenue and then placed in buckets based on the number of
elements that can fit in a bucket. The number of elements per bucket size is
defined as 8. The report includes the attribute Customer State and the
metrics Total Revenue and NTileSize. The syntax for the NTileSize metric is
as follows:
NTileSize ([Total Revenue],8)
The resulting report divides the 48 states into 6 buckets each containing 8
elements (states). A portion of the resulting report displays as follows.

NTileValue
NTileValue distributes the values in the metric by value ranges over a
user-defined number of buckets, sorted in either ascending or descending
order. Each bucket is the same size in terms of the range of values
contained in the bucket, but the number of elements per bucket is not equal.
This is an OLAP function.
Syntax
NTileValue <Ascending, Segments, BreakBy> (Argument)
Where:


l Segments is a positive integer designating the number of buckets in which

the values are distributed.
l Argument is a metric representing a list of values to be distributed into

buckets.
Example
Example 1: If you define the Segments=4, and the minimum value in the
range is 5 and the maximum is 105, the bucket distribution is as follows,
where x is the value:
l 5 £ x < 30 in bucket 1
l 80 £ x £ 105 in bucket 4
Example 2: This example shows a report where the customer states are
sorted based on revenue and then placed in buckets based on the value
range to which they belong. The number of buckets (segments) is defined as
4. The range of values is divided by the number of buckets and the result is
used to define four value ranges into which all the values fall. The report
includes the attribute Customer State and the metrics Total Revenue and
NTileValue. The syntax for the NTileValue metric is as follows:
NTileValue<Segments=4>([Total Revenue])
The resulting report displays the 48 states distributed across 4 buckets.

Notice that the number of elements in each bucket is not equal as more
values fall into one range than another. A portion of the resulting report is
displayed as follows.

Exam ple 3: Histogram
The reports and components used in this example are available in the
MicroStrategy Tutorial under the following folder:
MicroStrategy Tutorial\Public Objects\Reports\MicroStrate
gy Platform Capabilities\Advanced Analytics\Statistics an
d Forecasting\Histogram
How are My Customers Distributed (Classified) Based on Sales Data?
You are interested in finding out how your customers are distributed based
on their contributions to sales data. This example segments customers into
10 separate groups, with each group representing the customers within a
10% increment of sales. For example, the first segment includes the
customers who spend in the lowest 10% of the sales, while the final, tenth
segment includes the customers who spend in the top 10% of the sales.
This example uses the following concepts:

l Analytical functions: NTileValue
l Custom group
NTileValue distributes values into buckets based on minimum and maximum

values. These tiles are assigned an integer and the contents of each bucket
denoted accordingly. For example, all customers that fall in the first bucket
are assigned a 1, all customers in the second bucket are assigned a 2, and
so on. This function is computed by Intelligence Server. The metric using
NTileValue is then used to create a dynamic classification of Customers
using a custom group.
To create a general solution that can be used by different users, use object
prompts (see instructions below).
Create the following object prompts:
l Choose a base fact. This object prompt asks the user to select a fact:
Revenue, Profit, or Units Sold.
l Choose a sample set level. This object prompt asks the user to select an
attribute: Customer, Item, or Day.

Once you have identified the basic fact and attribute to analyze, you can
start building other objects. Take the following steps:
1. Create a metric defined as follows:
Sample Set Metric = Sum(?[Choose a base fact]){~, ?

[Choose a sample set level]}
2. To put the value in the proper buckets (tiles), create another metric
defined as:
Decile By Value = NTileValue<Segments=10> ([Sample Set

Metric])
3. Create a dynamic classification of the subject attribute (Customers,

Item, or Day) by building a custom group using banding. Use the metric
Decile By Value, the banding type "band count" and set the band count
to 10, starting at 1 and stopping at 10.

4. Choose to show only individual items within this element since the
objective is to build a graph (histogram) for this custom group.
5. Count the number of elements in the new classification to display a

histogram. To do this, create a dummy metric defined as follows:
Count of Samples = Count(1){~}

6. Since there is no column to perform the count, define the metric so that
the database can calculate how many attribute elements are in each
custom group element.
7. Add the Decile by Value custom group to the row axis, and the Count of
Samples metric to the column axis.
8. Execute the report with attribute Customer and fact Revenue, and the
Decile Histogram, in Grid mode, displays as follows:
The same report, in graph mode, appears as follows:

The SQL generated for the Decile by Histogram is as follows:

create table ZZT1Y02011CMQ000 (
CUSTOMER_ID SHORT,
DA56 LONG)
[An Analytical SQL]
insert into ZZT1Y02011CMQ000 values (1499, 1)
select a11.[DA56] AS DA56,
count(1.0) AS WJXBFS1
from [ZZT1Y02011CMQ000] a11
group by a11.[DA56]
drop table ZZT1Y02011CMQ000
l Pass0 computes the revenue for each Customer using the Sample Set
Metric.
l The next three passes (Pass1, Pass2, and Pass3) prepare the dynamic
classification of Customer for the custom group. This series of SQL passes
essentially builds a dynamic lookup table for the classification that is
requested. The column DA stands for Dynamic Attribute. The Intelligence
Server computes the NTileValue function. This is indicated by [An
Analytical SQL] in Pass3. The value is then inserted, with the new
classification, into the dynamic lookup table for the custom group.
l Pass4 calculates the number of customers that belong to each custom

group element.
l Pass5 drops the temporary table.

NTileValueSize
NTileValueSize distributes the values in the metric across buckets based
on a user-specified value range, sorted in either ascending or descending
order. This is an OLAP function.
Syntax
NTileValueSize <Ascending, BreakBy> (Argument, Size)
Where:

l Argument is a metric representing the list of values to be distributed

across buckets.
l Size is a real number designating the size of the range of values for each
bucket.
Example
Example 1: If you define size as 1000 and your range of values begins at
1200, the first bucket contains values 1200 through 2199, the second bucket
contains values 2200 through 3199, and so on until every value is in a
bucket. It does not matter how many values are in each bucket; it only
matters that their value is within the bucket range.
Example 2: This example shows a report where the customer states were
sorted based upon revenue and then placed in buckets based upon the value
range to which they belong. The value range is defined as 100,000.
Beginning at the lowest metric value, in this case 18,654, the first bucket
contains values from 18,654 to 118,653, the next bucket ranges from
118,654 to 218,653, and so on until all values are in buckets. The highest

value in the report is 1,839,238, which falls into bucket 19 and is the only
value in that range.
The report includes the attribute Customer State and the metrics Total
Revenue and NTileValueSize. The syntax for the NTileValueSize metric is
as follows:
NTileValueSize ([Total Revenue], 100000)
A portion of the resulting report is displayed as follows.
Percentile
Returns the n th percentile of values in a given range. This function can be
used to establish thresholds indicating, for example, which states have
revenue above the 75th percentile for their region. This is a group-value
function.
Syntax
Percentile <Ascending, FactID> (Argument, Percent)

Where:
l Ascending is a TRUE/FALSE parameter that indicates the organization of

the data.
l Percent is the decimal value of the percent for which you want to use as
a threshold.
Example
Example 1: If you have an argument containing the values 1, 2, 3, and 4 and

you want to find the threshold of the 3oth percentile, the syntax is as follows:
Percentile ({1, 2, 3, 4}, 0.3) = 1.9
1.9 is the level of the 30th percentile. All values above 1.9 are greater than
the 30th percentile.
Example 2: This example shows a report where only states with Total
Revenue greater than the threshold for the 75th percentile in its region are
displayed. The report contains the attributes Customer Region and
Customer State, and the metrics Total Revenue and Percentile. A view filter
is then applied to the report so that it only displays states where Total
Revenue is greater than Percentile.
The syntax for the Percentile metric is as follows:
Percentile([Total Revenue], .75){[Customer Region]}

PercentRank
This function is used to display the ranking of values as a percentage. For
example you can rank profit for a particular region based on the profit for all
regions.
The calculation can restart based on attributes identified in the parameter

You can also rank values in the following ways:
l You use the Rank function (see Rank , page 289) to rank values as an
integer value. For example you can rank profit for five categories from 1 to
5.
l You use the PercentRankRelative function (see PercentRankRelative,

page 286) to rank values as a percentage, with the ranking based on
another data set. For example you can rank profit for the current year
based on last year's profit.
Syntax
PercentRank <inclusive, BreakBy> (Argument)
Where:

l Argument is a fact or metric representing a list of numbers that are to be

ranked.
l inclusive is a TRUE/FALSE parameter that indicates whether the rank is

inclusive or exclusive:
l TRUE (default): The percent rank is in a range from 0 to 1 inclusive. This

means that 0% and 100% are included.
l FALSE: The percent ranks is in a range from 0 to 1 exclusive. This

means that 0% and 100% are excluded.
l BreakBy is the parameter that designates where the calculation should

restart.
Example
The example report shown below displays revenue information across

regions for a given quarter. You can use PercentRank to return the rank of
revenue for each region during a given quarter.
The definition for the Percent Rank Revenue metric in this example is:
PercentRank<BreakBy={Quarter}>(Revenue)
The Percent Rank Revenue metric in the report shown above provides the
rank of each revenue value. You can see that the highest value in the
Northeast region is displayed as 100% while the lowest value in the
Northwest region is displayed as 0%.

This report also includes a metric that uses the PercentRankRelative

function, ranking the revenue based on last quarter's revenue. For
information on PercentRankRelative and an explanation of these
results, see PercentRankRelative, page 286.
PercentRankRelative
This function is used to display the ranking of values as a percentage, with
the ranking based on a secondary data set. For example you can rank the
current year's profit for a particular region based on (relative to) the profit
from the previous year.
The calculation can restart based on attributes identified in the parameter

You can also rank values in the following ways:
l The Rank function (see Rank , page 289) ranks values as an integer value.
For example, you can rank profit for five categories from 1 to 5.
l The PercentRank function (see PercentRank , page 284) ranks values as

a percentage based on those values as the data set. For example, you can
rank profit for a particular region based on the profit for all regions.
Syntax
PercentRankRelative <inclusive, rankOutliers, significance,

truncate, BreakBy> (Argument, Rank)
Where:
l Argument is a fact or metric representing a list of numbers that are to be

ranked.
l Rank is a fact or metric representing a list of numbers that are used to

determine the rank of the values provided with Argument. The values for
Rank should be within the same range as the values of Argument to
provide a relevant ranking. For example, using profit values to rank
revenue would rarely result in relevant results, as profit is likely to have a

different range of values than Revenue. However, using last year's profit
to rank this year's profit can provide relevant results as the profits
between last year and this year are more likely to be in the same range of
values.
l inclusive is a TRUE/FALSE parameter that indicates whether the rank is

inclusive or exclusive:
l TRUE (default): The percent rank is in a range from 0 to 1 inclusive. This

means that 0% and 100% are included.
l FALSE: The percent rank is in a range from 0 to 1 exclusive. This means

that 0% and 100% are excluded.
l rankOutliers is a TRUE/FALSE parameter that indicates whether outlier

values are included in the calculation:
l TRUE: Values from Rank that are outside of the range of values from
Argument are included in the ranking. This often results in showing
values such as 100% or 0%.
l FALSE (default): Values from Rank that are outside of the range of
values from Argument are not included in the ranking. The results are
left blank rather than showing a percentage.
l significance (default is 3) determines the number of digits that are

used to perform each calculation. You can provide any integer value from
1 to 9 for this parameter.
l truncate is a TRUE/FALSE parameter that indicates whether the final

result is rounded or truncated:
l TRUE (default): The final result is truncated, based on the significance

applied to the calculation. For example, using the default of
significance=3, the fourth digit is dropped and is not used to round
the third digit.

l FALSE: The final result is rounded, based on the significance applied to

the calculation. For example, using the default of significance=3, the
fourth digit is used to round the third digit.

restart.
Example
The example report shown below displays revenue information across

regions for a given quarter. You can use PercentRankRelative to return
the rank of revenue for each region during a given quarter, relative to last
quarter's revenue.
The definition for the Percent Rank Revenue metric in this example is:
PercentRankRelative<BreakBy={Quarter}>(Revenue, [Last Quarter's

Revenue])
The Percent Rank Relative Revenue metric in the report shown above
provides the rank of each revenue value. The highest value in the Northeast
region displays no data because last quarter's revenue is $797,627, which is
greater than any revenue value for this quarter and thus outside of the range
of values. Since the default for the function is to not rank outlier values, no
data is displayed. If you modify the rankOutlier parameter to be
rankOutlier=TRUE, then data is returned for this ranking as shown in the
report below.

This report also includes a metric that uses the PercentRank function,
ranking the revenue based on its own values. For information on
PercentRank and an explanation of these results, see PercentRank , page
284.
Rank
This function is used to display the ranking of values in a list relative to the
other values. The calculation can restart based on attributes identified in the
parameter settings. This is an OLAP function.
Unless the defaults are changed, the function ranks the values in ascending
order by the value of the metric, and the rank is an integer.
Syntax
Rank <ASC, ByValue, BreakBy, NullInclude> (Argument)
Where:
l Argument is a fact or metric representing a list numbers.
l ASC is a TRUE/FALSE parameter that indicates the order of ranking (1 is

the lowest or highest value).
l ByValue is a TRUE/FALSE parameter that indicates whether the ranking is

done by integer values (1, 2, 3, 4) or by percentage (10%, 50%, 75%,
100%).

restart.

l NullInclude is a parameter that determines how NULL values are

included in the rank calculation.
The NullInclude parameter only affects the rank of NULL values if the
Rank function is performed by the MicroStrategy Analytical Engine. The
Rank function is performed by the Analytical Engine for smart metrics,
derived metrics, and other metric scenarios. To determine whether the
Rank function for a metric is performed by the Analytical Engine, view the
SQL statement for the report. If the metric is listed in the Analytical Engine
calculation steps, this verifies that the Rank function is performed by the
Analytical Engine.
If the Rank function is performed on a database, the NullInclude

parameter is ignored and NULL values are included in the rank calculation
based on the database standards.
For Rank functions that are performed by the Analytical Engine, you have
the following options for this parameter:
l 1: If you define NullInclude=1, NULL values are given a rank value

equal to the number of other rank values, plus one. For example, the
rank of the Profit metric in the report below ranks four separate profit
values.
There is one NULL value, which is given the rank of 4. The Rank (Profit)
metric in this example uses an ascending rank. If you define the metric
with a descending rank, the other rank values change but the rank value
for any NULL values remains the same. This is shown in the report

below.
l -1: If you define NullInclude=-1, NULL values are given the rank
value of one. For example, the reports shown below both define the
Rank (Profit) metric with NullInclude=-1. The report on the left uses
an ascending rank, while the report on the right uses a descending rank.
As shown in the reports above, the NULL values for both reports are
ranked with the value of one.
l 0 (default): If you define NullInclude=0, NULL values are included in

the rank calculation based on the NULL value handling defined using the
Null checking for Analytical Engine VLDB property. For information on
VLDB properties, including steps to access and modify them for various
MicroStrategy objects, see the System Administration Guide :
If you define the Null checking for Analytical Engine property as True,
NULL values are treated as zero values in the rank calculation. For
example, the report shown below ranks the NULL values with a rank of
two, because zero is greater than -10 and less than 40.

If you define the Null checking for Analytical Engine property as False,
NULL values are treated as NULL values, which means NULL values are
also displayed for the rank values. For example, the report shown below
displays the NULL values as NULL in the rank.
Example
This example report displays customer states ranked by revenue within their
regions. There are two metrics, one that ranks by value (default) and the
other that ranks using a percentage. In order to have the state with the
highest revenue in each region ranked as 1 in the Rank by Value metric, the
Rank by Value ranking is descending. In order to have the largest number
ranked at 100% in the Rank by Percent metric, the ranking is ascending
(default).
The report includes the attributes Customer Region, Customer State, the
metric Total Revenue (defined as Sum(Revenue)), and the two ranking
metrics. The syntax for the metrics is as follows:
Rank by Value:
Rank<ASC=False,ByValue=True, BreakBy={[Customer Region]}>([Total

Revenue])
Rank by Percent:

Rank<ASC=True,ByValue=False BreakBy={[Customer Region]}>([Total

Revenue])
The resulting report is displayed as follows.
String Functions
String functions perform various actions that modify the characters returned
for a string of characters. While string functions can be used to create
metrics, a more common use case for these functions is in the creation of
attribute forms. For example, these functions can aid in the creation of
attribute forms by combining multiple columns of information, capitalizing
the first letter of a column, removing or returning select characters of a
column, and so on. For information on creating attributes and attribute
forms, see the Project Design Help.
The MicroStrategy Analytical Engine does not calculate string functions;

they are processed by the database. For information on which string
functions are supported for your specific database, see Appendix A,
MicroStrategy and Database Support for Functions, and search in the
section corresponding to your database. String functions for which your
database does not include SQL syntax support cannot be calculated in your
environment.

This section of the document includes information and examples on the data
returned by the function.
BeginsWith
Returns 1 if a text string begins with a specified text pattern. If the text string
does not begin with the pattern, the function returns 0.
For the definition and syntax of the Begins with comparison operator, see
Begins With, page 326.
Syntax
BeginsWith(String, Pattern)
Where:
l String is the string that is being searched. You can use facts, metrics,
columns, or string values.
l Pattern is the string that is being searched for at the beginning of the
values from String. You can use facts, metrics, columns, or string
values.
The text pattern comparison may or may not be case sensitive depending on
the database implementation.
Example
BeginsWith(Region@DESC, "North")
Returns 1 for Regions that start with North, such as Northeast and
Northwest.
Char (Convert ASCII Code to a Character)

The Char function converts a decimal ASCII code into its associated
character. This function supports returning any of the standard 128
characters assigned an ASCII code.

Syntax
Char(Argument)
Where:
Argument is a fact, metric, column, or constant value that provides an

ASCII code in ASCII decimal format. Any value provided outside the range of
0 to 127 causes the Char function to return a single question mark (?)
character. Any other invalid input, such as character strings, causes the
Char function to return null.
For information on the syntax used in your specific database, see Appendix
A, MicroStrategy and Database Support for Functions and see the section
that corresponds to your database.
Example
Char(65)
This example returns A.
Concat (Concatenate)
The Concat function combines two or more input strings into one
continuous string and returns the result. For example, concatenating the two
strings "Micro" and "Strategy" would result in the single string
"MicroStrategy". Concat is often used to combine related values, such as a
first and last name.
Syntax
Concat(Argument1, Argument2,..., ArgumentN)
Where:
Argument1,..., ArgumentN are facts, metrics, columns, or string

values.

Example
The Concat function can help to create attribute forms that are a
combination of multiple columns in database tables. For example, an LU_
CUSTOMER table includes two columns: CUST_FIRST_NAME and CUST_
LAST_NAME. You can create a single attribute form that combines these two
columns as described below.
Concat([CUST_FIRST_NAME], [CUST_LAST_NAME])
The input from the first column is concatenated with the input from the
second column, to display information such as JohnDoe and JaneDoe for the
attribute form.
For scenarios such as the one described above, you can use the
ConcatBlank function (see ConcatBlank (Concatenate Plus Blank Space),
page 297 below) to concatenate the strings and include a space between
the two strings. This can result in attribute forms that are easier to read. For
example, rather than displaying JohnDoe, the attribute form would display
John Doe.
For an additional example of using the Concat function, see the example
section for the function InitCap (Initial Capitalization), page 299.
ConcatAgg (Concatenate Plus Delimiter)

The ConcatAgg function takes all of the content from a single input and
concatenates the content as a single string. By default, each string of
characters that is concatenated is separated by a comma. You can define
the delimiter.
Syntax
ConcatAgg<Delimiter, FactID

Where:
l Argument is facts, metrics, columns, or string values.
l Delimiter is a parameter that determines the characters used to

separate each concatenated value. By default, a comma is used.

Example
The ConcatAggBlank function can help to create attribute forms that are a
combination of multiple rows in database tables. For example, an LU_
CUSTOMER table includes a column CUST_LAST_NAME. You can create a
single attribute form that combines all the values (rows) for CUST_LAST_
NAME into a single attribute form:
ConcatAgg([CUST_LAST_NAME])
Each last name is concatenated, separating each last name with a comma
by default.
ConcatBlank (Concatenate Plus Blank Space)

The ConcatBlank function concatenates two or more input strings into one
continuous string, inserting a blank space between each string. This function
can be used to combine related values that are logically separated by
spaces. For example, concatenating the two strings "Business" and
"Intelligence" would result in the single string "Business Intelligence".
Syntax
ConcatBlank (Argument1, Argument2,..., ArgumentN)
Where:

Argument1,..., ArgumentN are facts, metrics, columns, or string

values.
Example
The ConcatBlank function can help to create attribute forms that are a
combination of multiple columns in database tables. For example, an LU_
CUSTOMER table includes two columns: CUST_FIRST_NAME and CUST_
LAST_NAME. You can create a single attribute form that combines these two
columns as described below.
ConcatBlank([CUST_FIRST_NAME], [CUST_LAST_NAME])
The input from the first column is concatenated with the input from the
second column, and an additional space is included between the two inputs.
This displays information such as John Doe and Jane Doe for the attribute
form.
For an additional example of using the ConcatBlank function, see the

example section for the function InitCap (Initial Capitalization), page 299.
EndsWith
Returns 1 if a text string ends with a specified text pattern. If the text string
does not end with the pattern, the function returns 0.
For the definition and syntax of the Ends With comparison operator, see
Ends With, page 328.
Syntax
EndsWith(String, Pattern)
Where:

l String is the string that is being searched. You can use facts, metrics,
columns, or string values.
l Pattern is the string that is being searched for at the end of the values
from String. You can use facts, metrics, columns, or string values.
Example
EndsWith ([Unit Price], ".99")
Returns 1 for prices that end in .99, such as 10.99, 11.99, and so on.
If String is an attribute form, the EndsWith function can only be used to create
attributes, not metrics. If you use an expression similar to the one below to create a
new metric, you receive an error message.
EndsWith(Employee@[Last Name], "son")
InitCap (Initial Capitalization)

The InitCap function returns a string in which the first letter of the input
string is capitalized. All other letters appear in lower case. This can help to
fix capitalization errors in information that is displayed for attribute forms,
metrics, and other objects.
To capitalize the first letter of every word in a string, see TitleCap (Title
Capitalization), page 314.
Syntax
InitCap(Argument)
Where:
Argument is a metric, column, or string value representing the text string.
For information on the syntax used in your specific database,see Appendix

Example

The InitCap function can be used to fix capitalization errors for information
that is displayed for attribute forms. For example, an LU_CUSTOMER table
includes two columns: CUST_FIRST_NAME and CUST_LAST_NAME. To
ensure that the first letter of the last name is capitalized you create an
attribute form with the following definition.
InitCap([CUST_LAST_NAME])
The input from the column is modified so that the first character is
capitalized and all other characters are lowercase. For example, if the
column included information such as jackson, sMITh, and Hughes, these
would be displayed as Jackson, Smith, and Hughes respectively.
Be aware that this function could potentially return undesired results in

certain scenarios. Using the scenario described above, consider the name
McCoy. Using the InitCap function, this would be displayed as Mccoy.
Another way to use this function would be to create an attribute form that
combined the first letter of someone's first name with the person's full last
name. The InitCap function could be used to ensure that the first letter of
the first name was capitalized. An attribute form of this type would require
the use of the functions InitCap, ConcatBlank, Concat, and LeftStr.
The definition of such an attribute form is shown below:
ConcatBlank(Concat(LeftStr(InitCap([CUST_FIRST_NAME]), 1), "."),

[CUST_LAST_NAME])
An attribute form using a definition such as the one listed above would
display names such as J. Doe, M. Smith, and L. Martinez. The InitCap
function ensures that the first letter of the first name is capitalized.
LastPosition (Last Position of Substring)

The LastPosition function returns the starting position of the last
occurrence of a series of characters in the input string. For example, using
the LastPosition function to search for the string "Strategy" within the
string "MicroStrategy Inc. MicroStrategy" would return the value of 25.

To find the first occurrence of a series of characters in a string, see Position

(Position of Substring), page 307.
Syntax
LastPosition (Argument1, Argument2)
Where:
l Argument1 is the string in which to search for Argument2.
l Argument2 is the substring to search for.
LeftStr (Left String Selection)

The LeftStr function returns a substring taken as a specified number of
characters from the left of the input string. For example, if the specified
length is five, LeftStr would return the string "Micro" from the original
string "MicroStrategy". LeftStr is useful for abbreviations or length
reduction when the entire input string is not required.
Syntax
LeftStr (Argument, Length)
Where:
l Argument is a metric, fact, column, or string value representing the text

string(s).
l Length is an integer indicating the number of characters, starting from the

far left position of the string, to be returned.
Example

The LeftStr function can be used to create attribute forms that use
abbreviations. This can help reduce the length for attribute forms that may
have long descriptions.
For example, an LU_CUSTOMER table includes two columns: CUST_FIRST_

NAME and CUST_LAST_NAME. You create attribute forms for the customer
first and last names. You also create an attribute form that displays only the
first letter of the first name. This attribute form can be used when the full
first name does not need to be displayed. The definition of such an attribute
form is shown below:
LeftStr([CUST_FIRST_NAME], 1)
The integer value of 1 causes the LeftStr function to only display the first
character at the far left position for all inputs from the CUST_FIRST_NAME
column.
For an additional example of using the LeftStr function, see the example
section for the function InitCap (Initial Capitalization), page 299.
Length (Length of String)

The Length function returns the number of characters in an input string. For
example, using the Length function on the string "MicroStrategy" would
return a value of 13. Length is often used to manipulate strings with the
help of other string functions.
Syntax
Length(Argument)
Where:
Argument is a metric, fact, column, or string representing the text string(s).
For information on the syntax used in your specific database, w33 Appendix
Example

The Length function can be used in conjunction with other string functions
to manipulate strings in various ways. For example, rather than abbreviating
information to a specific number of characters, you can abbreviate each
string to only display half of the available characters. The definition of an
attribute form that displays only half of the characters available for the first
names of customers is shown below:
LeftStr([CUST_FIRST_NAME], Int(Length([CUST_FIRST_NAME]) / 2))
For first names such as Alan, Frederick, and Jennifer, the function shown
above would display Al, Fred, and Jenn, respectively. This provides an
abbreviated version of customers' first names while providing additional
characters to possibly distinguish each abbreviated name.
In the function used in this example, the Int function is used to return only
the integer part of the calculation Length([CUST_FIRST_NAME]) / 2.
Depending on your database's support for functions, you could also use
functions such as Round, Ceiling, and Floor. For information on these
types of functions, see Mathematical Functions, page 415.
Lower (Lower Case)

The Lower function returns a string in which all alphabetic characters in an
input string are displayed as lower case. For example, using the Lower
function on the string "MicroStrategy" would return the string
"microstrategy".
Lower may be used to help standardize the display of information and make
the information more readable.
Syntax
Lower(Argument)
Where:
Argument is a metric, column, or string value representing the text string

(s).

Example
The Lower function can be used to display information in all lower case for
attribute forms. For example, an LU_CUSTOMER table includes an ADDRESS
column. You can create an attribute form to display all characters for a
customer's address in lower case. The definition of such an attribute form is
shown below:
Lower([ADDRESS])
An attribute form using a definition such as the one listed above would take
input addresses such as 10 Main Street, 350 West Elm Avenue, and 4400
Spring Road and display them as 10 main street, 350 west elm avenue, and
4400 spring road, respectively.
LTrim (Left Trim)

The LTrim function returns a string in which any leading blank spaces on
the left side of the input string have been removed. For example, using the
LTrim function on the string " MicroStrategy" would return the string
"MicroStrategy".
The LTrim function helps to remove blank spaces that may have been
caused by errors in data entry. Removing these spaces helps standardize
the display of information and makes the information more readable.
Syntax
LTrim(Argument)
Where:

Example
An LU_CUSTOMER table includes an ADDRESS column. You can create an

attribute form to remove any leading blank spaces from the addresses. The
definition of such an attribute form is shown below:
LTrim([ADDRESS])
Any leading blank spaces included for addresses are removed from the
display for the attribute form.
To remove both leading and trailing blank spaces, use the Trim function
(Trim, page 316).
Match
The Match function uses regular expressions to search a string for a pattern
of characters and returns any matches that are found.
Syntax
Match <Group=0,Instance=1>(Argument, Find)
Where:
l Argument is a metric, fact, column, or string representing the text strings

that are searched for matches.
l Find is a metric, fact, column, or string that provides a regular expression

used to searched the strings returned by Argument. The regular
expressions supported by this function conform to the standards of the
International Components for Unicode. For information about these
regular expression standards and syntax, see http://userguide.icu-
project.org/strings/regexp.

l Group is a parameter that determines which group within the regular

expression is returned. The groups in a regular expression are each set of
characters enclosed by parentheses () and are ordered left to right. By
default, the Group parameter is defined as 0 and the entire string that is
matched is returned. If you define Group as 1, only the first group from the
left of the regular expression is returned. If you define Group to a value
greater than the number of groups in the regular expression, no results are
returned. For example if there is one group but Group is defined as 2, no
results are returned.
l Instance is a parameter that determines which instance of the matching

results are returned. By default, Instance is defined as 1 and the first
match is returned. If you define Instance to a value greater than the
number of matching results, no results are returned. For example if there
are two matches but Instance is defined as 3, no results are returned.
For information about the syntax used in your specific database, the
Appendix A, MicroStrategy and Database Support for Functions and see the
section that corresponds to your database.
Example
Consider the string of text "Telephone: 703-555-1234, Fax: 704-555-6789".

You can search for and return various parts of the telephone numbers in this
string using regular expressions. The regular expression (\d+)[-\b]
(\d+)[-\b](\d+) breaks up the three sets of digits in a phone number
into separate groups, which can be returned using the Match function.
See http://userguide.icu-project.org/strings/regexp for information on the

regular expression standards and syntax.
For example:
l Match("Telephone: 703-555-1234, Fax: 704-555-6789", "(\d+)[-\b]

(\d+)[-\b](\d+)")
By default this returns the entire string that is the first match, which is 703-555-1234.

l Match<Group=1>("Telephone: 703-555-1234, Fax: 704-555-6789", "

(\d+)[-\b](\d+)[-\b](\d+)")
This returns the first group (the first search criteria in parentheses) from the first
occurrence, which is 703.
l Match<Instance=2, Group=3>("Telephone: 703-555-1234, Fax: 704-

555-6789", "(\d+)[-\b](\d+)[-\b](\d+)")
This returns the third group (the third search criteria in parentheses) from the second
occurrence, which is 6789.
Position (Position of Substring)

The Position function returns the starting position of the first occurrence
of a series of characters in the input string. For example, using the
Position function to search for the string "Strategy" within the string
"MicroStrategy" would return the value of 6.
To find the last occurrence of a series of characters in a string, see

LastPosition (Last Position of Substring), page 300.
In addition to locating the position of a substring, Position also can be

used to test whether the substring is present. If a given substring is not
found, 0 will be returned.
Syntax
Position (Argument1, Argument2)
Where:
l Argument1 is the substring to search for.
l Argument2 is the string in which to search for Argument1.
Example

The Position function can be used in conjunction with other functions to

perform various manipulations on strings. For an example of how this
function works, consider an LU_CUSTOMER table that includes an ADDRESS
column. You use the Position function to find the word Street in the
addresses as defined below.
Position("Street", [ADDRESS])
For an address of 10 Main Street, the function listed above returns the value
of 9. For addresses that do not include the word Street, a value of 0 would
be returned.
RepeatStr (Repeat String)

The RepeatStr function returns a character or string of characters the
specified number of times.
Syntax
RepeatStr(String, Times)
Where:
l String is a metric, fact, column, or string of characters representing the

characters to be repeated.
l Times is a metric, fact, column, or constant value that specifies how many
times to repeat the string of characters. A common practice is to type a
constant value.
Example
RepeatStr("Hello ", 3)
This would return the string of characters "Hello Hello Hello ".
Replace
The Replace function searches a string for a pattern of characters and
replaces each instance of those characters with the new characters you
specify. The resulting string with all instances replaced is returned.

Syntax
Replace(Argument, Find, Replace)
Where:
l Argument is a metric, fact, column, or string representing the text strings

that are searched.
l Find is a metric, fact, column, or string representing the text strings that
are searched for within the strings returned by Argument.
l Replace is a metric, fact, column, or string representing the text

strings that are used to replace any instances of the string from Find
that are included in the strings returned by Argument.
RightStr (Right String Selection)

The RightStr function returns a substring taken as a specified number of
characters from the right of the input string. For example, the first eight
characters from the right of the string "MicroStrategy" would be "Strategy".
RightStr can be used to create attribute forms that display only part of the
information available. This can be helpful when some information needs to
be hidden for security purposes, as in the credit card example below.
Syntax
RightStr (Argument, Length)
Where:
l Argument is a metric, fact, column, or string representing the text string

(s).
l Length is an integer indicating the number of characters to be returned.

Example
An LU_CUSTOMER table includes a CREDIT_CARD_NUMBER column that

contains credit card numbers for online customers. You create attribute
forms to display only the last four digits of the customers' credit card
numbers. The definition of such an attribute form is shown below:
RightStr([CREDIT_CARD_NUMBER], 4)
The integer value of 4 causes the RightStr function to only display the last
four characters from the CREDIT_CARD_NUMBER column.
You could also combine this functionality with the Concat function to
display Xs or other characters to represent the digits for the credit card that
are not displayed.
Concat("XXXX-XXXX-XXXX-", RightStr([CREDIT_CARD_NUMBER], 4))
This definition would modify the display of a credit card number from 1111-
2222-3333-4444 to be displayed as XXXX-XXXX-XXXX-4444.
RTrim (Right Trim)

The RTrim function returns a string in which blank spaces on the right side
of the input string have been removed. For example, using the RTrim
function on the string "MicroStrategy " would return the string
"MicroStrategy".
RTrim helps to remove trailing blank spaces, which may have been caused
by errors in data entry. Removing these spaces helps standardize the
display of information and makes the information more readable.
Syntax
RTrim(Argument)
Where:

Example

attribute form to remove any trailing blank spaces from the addresses. The
definition of such an attribute form is shown below:
RTrim([ADDRESS])
Any trailing blank spaces included in the addresses are removed from the
display for the attribute form.
To remove both leading and trailing blank spaces, use the Trim function
(Trim, page 316).
Split
The Split function searches a string, separates the contents into groups of
characters based on a delimiter, and returns the string of characters
requested. For example, you can search a string of characters that
separates values with commas, and return the characters after the first
comma. Searching the string "red, yellow, green" and selecting to return the
second set of characters based on a comma as the delimiter returns "yellow
".
Syntax
Split<SeparatorRegex=False>(String, Delimiter, Index)
Where:
l String is a metric, fact, column, or string representing the text strings

that are searched.

l Delimiter is a metric, fact, column, or string that determines the

characters used to separate each concatenated value. A common practice
is to provide a constant text value for the delimiter. Typing "," uses a
comma as the delimiter.
You can also choose to use regular expressions to define the delimiter
characters. For information on supporting regular expressions, see the
SeparatorRegex parameter described below.
If the delimiter character is not found in the string of characters, one of two
resolutions can occur:
l If you define Index as 1, then the entire string of characters is returned.
l If you define Index as any value other than 1, a NULL value is returned.
l Index is a metric, fact, column, or constant value that determines which

group of characters are returned. A common practice is to provide a
constant numerical value. For example, using a value of 1 returns the first
group of characters before the first delimiter characters are found in the
string.
l SeparatatorRegex is a parameter that determines if a regular

expression is used to provide the delimiter characters. The regular
expressions supported by this function conform to the standards of the
Intern a ion al Components for Unicode. For information about these
regular expression standards and syntax, see http://userguide.icu-
project.org/strings/regexp.
Example
l Split<SeparatorRegex=True>("red yellow green", "\s", 2)
This returns the string of characters "yellow". By using the regular expression \s, any
white space character is considered a delimiter. This splits the string into the strings
"red", "yellow", and "green".
l Split("red; yellow; green", ";" 1)

This returns the string of characters "red".
This returns the string of characters " green".
This returns a NULL value, since the string is only split into three separate strings.
SubStr (Substring Selection)

The SubStr function returns a substring taken as a specified sequence of
characters from the input string. SubStr is useful for isolating a specific
section of a string that contains relevant information.
Syntax
SubStr(Argument, Position, Length)
Where:
l Argument is a metric, fact, column, or string representing the text string

(s).
l Position is an integer indicating the starting position inside the string.
l Length is an integer indicating the number of characters to be returned.
Example
An LU_ITEM table includes an ITEM_ID column. This column stores a ten-

digit number, which has been created to define various facts about the item.
This includes two digits which identify the location in which the item is sold.
These two digits always start at the fourth digit from the left side of the ten-
digit item ID number. To retrieve these digits and determine the location of

sale for an item, you can create an attribute form to return these two digits.
The definition of such an attribute form is shown below:
SubStr(ITEM_ID, 4, 2)
The substring retrieved starts at the fourth digit and retrieves two
characters. For an item ID of 2334560897, this function returns 45.
TitleCap (Title Capitalization)

The TitleCap function returns a string in which the first letter of every word
in the input string is capitalized. All other letters that are not the first letter in
a word appear in lower case. This can help to fix capitalization errors in
information that is displayed for attribute forms, metrics, and other objects.
A word in a string is any string of alphabetic characters that are separated

by a non-alphabetic character such as a space, comma, or number.
To capitalize only the first letter of the first word in a string, see InitCap
(Initial Capitalization), page 299.
Syntax
TitleCap(Argument)
Where:
Argument is a metric, column, or string value representing the text string.
Example
TitleCap("john smith")
This example would return John Smith.
TitleCap("john SMITH")
This example would also return John Smith.

ToNumber (Convert String to a Number)

The ToNumber function converts a string of characters to its applicable
numeric value. Converting data that has been processed as a string to a
numeric value allows you to perform various calculations and analysis that
can be done only with numeric values.
Syntax
ToNumber(Argument)
Where:
Argument is a fact, metric, column, or constant value that provides the

strings that are converted to numeric values. The function can recognize E
notation and return the appropriate number.
Example
ToNumber("1001")
This example returns the numeric value 1001.
ToNumber("1.48e12")
This example returns the numeric value 1,480,000,000,000.
ToString (Convert Number, Date, or Timestamp to a String)

The ToString function converts a number, date, or timestamp to a string of
characters. Converting data that has been processed as a numeric, date, or
timestamp value to a string allows you to view and display the data in
different ways.
Syntax
ToString<Pattern=null>(Argument)
Where:

l Argument is a fact, metric, column, or constant value that provides the

values that are converted to a string of characters.
l Pattern is a parameter that determines the formatting for the resulting

string of characters. When providing a pattern, enclose the pattern in
double quotes (""). For example:
ToString<pattern="0,000.00">(Revenue)
To specify a pattern, you can use the custom numeric, date, and time
formatting symbols that are described in the Advanced Reporting Help.
Example
ToString<pattern="0,000.00">(1001)
This example returns the string "1,001.00".
Trim
The Trim function returns a string in which blank spaces on either side of
the input string have been removed. For example, using the Trim function
on the string " MicroStrategy" would return the string "MicroStrategy".
The Trim function helps to remove leading and trailing blank spaces, which
may have been caused by errors in data entry. Removing these spaces
helps standardize the display of information and makes the information more
readable.
Syntax
Trim(Argument)
Where:
Example


attribute form to remove any leading and trailing blank spaces from the
addresses. The definition of such an attribute form is shown below:
Trim([ADDRESS])
Any leading or trailing blank spaces included in the addresses are removed
from the display for the attribute form.
Upper (Upper Case)

The Upper function returns a string in which all alphabetic characters in an
input string are displayed as upper case. For example, using the Upper
function on the string "MicroStrategy" would return the string
"MICROSTRATEGY". The Upper function can be used to display information
in all upper case for attribute forms.
Syntax
Upper(Argument)
Where:
Argument is a metric, column, or string representing the text string(s).
Example

attribute form to display all characters for a customer's address in upper
case. The definition of such an attribute form is shown below:
Upper([ADDRESS])
An attribute form using a definition such as the one listed above would take
input addresses such as 10 Main Street, 350 West Elm Avenue, and 4400

Spring Road and display them as 10 MAIN STREET, 350 WEST ELM
AVENUE, and 4400 SPRING ROAD, respectively.

O PERATORS

The operators described in this chapter are building block functions. They
provide the means to perform simple mathematical operations, compare
values, search strings, and evaluate logical conditions.
Operators may not be available for direct selection when creating

MicroStrategy objects such as metrics. These operators are employed to
support the logic of metrics, filters, thresholds, and so on.
The following categories of operators are covered in this chapter:
Arithmetic Operators
The arithmetic operators are basic mathematical functions, such as Minus,
Times, Divide, Plus, and Unary Minus, which are among the most commonly
used operators. A brief description of each operator follows.
Minus (-)
Returns the difference between two values.
Syntax
Arg1 - Arg2
Where:
Arg1 and Arg2 are attributes, facts, metrics representing a list of numbers,
big decimals, date/time, or functions that return numbers or big decimals.
Both Arg1 and Arg2 must be of the same data type, with the exception if
Arg1 is a Date data type. In this case, Arg2 must be a number.
Example
A metric is defined as:
Revenue - Freight
This metric returns the difference between the revenue and freight charges.

Times (*)
Returns the product of two values.
Syntax
Arg1 * Arg2
Where:
Arg1 and Arg2 are attributes, facts, metrics representing a list of numbers
or big decimals, or functions that return numbers or big decimals.
Example
([Unit Profit] * [Units Sold])
This metric returns the product of the unit profit and the units sold values.
Divide (/)
Returns the quotient when one value is divided by another.
Syntax
Arg1 / Arg2
Where:
or big decimals, or functions that return numbers or big decimals. In
addition, Arg2 must not be zero.
Example
(Profit / [Units Sold])
This metric returns the quotient of the profit and number of units sold.

Plus (+)
Returns the sum of two values.
Syntax
Arg1 + Arg2
Where:
or big decimals, or functions that return numbers and big decimals. In
addition, if either of the two arguments is date/time, the other argument must
be a number.
Example
Cost + Freight
This metric returns the sum of the cost and the freight charges.
Unary minus (U-)

Returns the absolute value of a negative value or the negative value of a
positive value.
Syntax
U-(Arg)
Where:
Arg is an attribute, fact, metric representing a list of numbers, or a function

that returns numbers and big decimals.
Example
U-(Profit)

This metric changes the sign of the profit value. If the profit value is
negative, it returns a positive value. If the profit value is positive, it returns a
negative value.
Comparison Operators
These operators are used to compare values, which can be numbers, text
strings, date and time, expressions, or operators that return any values of
the mentioned data types. A brief description of each operator follows. The
comparison patterns of text strings may or may not be case-sensitive
depending on the database implementation.
Less Than (<)

Returns TRUE if the first value is less than the second value.
Syntax
Arg1 < Arg2
Where:
text strings, date/time, or functions that return literal values. Arg1 and Arg2
must be of the same data type.
Example
A condition is defined as:
Profit < Cost
This condition returns TRUE only if the profit is less than the cost.
Less Than or Equal (<=)

Returns TRUE if the first value is less than or equal to the second value.
Syntax
Arg1 <= Arg2

Where:
Example
Profit <= Cost
This condition returns TRUE only if the profit is less than or equal to the
cost.
Not Equal (<>)

Returns TRUE if two given values are not equal to each other.
Syntax
Arg1 <> Arg2
Where:
Example
Profit <> Cost
This condition returns TRUE only if the profit is not equal to the cost.
Equal (=)
Returns TRUE if the two given values are equal to each other.
Syntax
Arg1 = Arg2

Where:
Example
Profit = Cost
This condition returns TRUE if the profit is equal to the cost.
Greater (>)
Returns TRUE if the first value is greater than the second value.
Syntax
Arg1 > Arg2
Where:
Example
Profit > Cost
This condition returns TRUE only if the profit is greater than the cost.
Greater Than or Equal (>=)

Returns TRUE if the first value is greater than or equal to the second value.
Syntax
Arg1 >= Arg2

Where:
Example
Profit >= Cost
This condition returns TRUE only if the profit is greater than or equal to the
cost.
Begins With
Returns TRUE if a text string begins with a specified text pattern.
For the definition and syntax of the BeginsWith string function, which can be
used to create attribute forms, see BeginsWith, page 294.
Syntax
Arg1 Begins With Arg2
Where:
Arg1 and Arg2 are text strings.
The text pattern comparison may or may not be case sensitive depending on
the database implementation.
Example
Region@DESC Begins with "North"
Returns TRUE for Regions that start with North, such as Northeast and
Northwest.

Between
Returns TRUE if the specified value is between the inclusive range of two
boundaries. Between can be used to test if a value is within a valid data
range.
Syntax
Value Between Boundary1 and Boundary2
Where:
Value, Boundary1, and Boundary2 can be numbers, date/time, text, or

functions that return the mentioned data types. Value, Boundary1 and
Boundary2 must be of the same data type.
Example
l M1 Between 1 and 1000
l region@DESC Between "A" and "Z"
l date@DESC Between '1999-2-12' and '2005-2-23'
Contains
Returns TRUE if a text string contains a specified text pattern.
Syntax
Arg1 Contains Arg2
Where:
Arg1 and Arg2 must be of data type Text.
Example
Employee@[Last Name] Contains "Smith"
Returns TRUE for Employees with a last name such as "Smith", "Smithson",
and so on.

Ends With
Returns TRUE if a text string ends with a specified text pattern.
For the definition and syntax of the EndsWith string function, which can be
used to create attribute forms, see EndsWith, page 298.
Syntax
Arg1 Ends With Arg2
Where:
Example
Employee@[Last Name] Ends with "son"
Returns TRUE for Employees with a last name that ends with son, such as
Wilson, Johnson, and so on.
In
Returns TRUE if a value is contained in a specified list of values.
Syntax
Arg1 In Arg2
Where:
Arg2 is a list of the literals with the same data type as Arg1. It can be of the
data type numbers, big decimals, text, date/time, or functions that return
these data types.
Examples
l Employee@[Last Name] in ("Smith", "Cooper", "Michael")
l Year in (2000, 2001, 2002)

Like
Returns TRUE if a text string matches a specified text pattern; otherwise,
returns FALSE.
Depending on whether you use wildcards and how they are used in the text
in the pattern, Like can be used in place of Begins With, Ends With,
Contains, or =. This operator is used to search for related strings.
Syntax
Arg1 Like Arg2
Where:
Usage Notes
Using wildcards with the Like operator allows you to search for more than
just a static set of text. For example, rather than searching for the exact text
pattern South, you can use wildcards to search for any text pattern that
includes South, such as Mid Southern, SouthEast, and South.
The Like operator can be processed by the MicroStrategy Analytical

Engine, or it can be passed to the database to be processed using the
database's own comparison support. You can determine if the Like operator
was processed by the MicroStrategy Analytical Engine or passed to the
database by viewing the SQL view of a report. If the Like operator is
included in the SQL of the report, the database performed the comparison.
By contrast, if the Like operator is not included in the SQL of the report, or
the Like operator is part of retrieving results from an Intelligent Cube (see
the In-memory Analytics Guide for reporting on Intelligent Cubes), then the
MicroStrategy Analytical Engine processed the Like operator. Wildcard
support depends on how the Like operator is processed:

l If the MicroStrategy Analytical Engine processes the Like operator, the

following wildcard characters are supported:
l The % character can be used to represent any number of characters. For

example, using the comparison Like 'Sout%' returns TRUE for South,
SouthEast, and Southern.
l The _ character can be used to represent a single character. For

example, using the comparison Like 'Sout_' returns TRUE for South
but returns FALSE for SouthEast and Southern.
l The \ character can be used as an escape character for the %, _, or \

characters, which means it can be used prior to these wildcard
characters to search for the character rather than to use it as a wildcard.
To search for the characters %, _, or \, you must include a single \
character before the character you are searching for. For example, to
search for the exact text User_ID, you would need to use the
comparison Like 'User\_ID'.
l The * character can be used to represent any number of characters. For

example, using the comparison Like 'Sout*' returns TRUE for South,
SouthEast, and Southern. The * character also acts as its own escape
character, which means it can be used prior to another * to search for
asterisks in text patterns. For example, to search for the exact text
User*, you would need to use the comparison Like 'User**'.
l If the database processes the Like operator, the database determines

how wildcards are supported. In general, many databases support the
same wildcard characters and escape characters that are listed above as
supported by the MicroStrategy Analytical Engine. However, some
databases do not support \ as an escape character and instead use an
alternative such as enclosing the wildcard character in brackets. For
example, a database may support using [%] to search for the % character.
Therefore, if the Like operator is being processed by the database, see
your third-party database documentation to verify which wildcards can be
used as part of a comparison.

Example
l Region@DESC like 'South'
Returns TRUE if region is "South".
l Region@DESC like 'South%'
Returns TRUE if region is "South," "SouthEast," and so on.
l Region@DESC like '%South%'
Returns TRUE if region is "Mid Southern," "SouthEast," "South," and so on.
l Region@DESC like 'D_g'
Returns TRUE if region is "Dog," "Dig," "Dug," and so on.
l Region@DESC like 'D*g'
Returns TRUE if region is "Dog," "Drag," "Drug," and so on.
Not Begins With

Returns TRUE if a text string does not begin with a specified text pattern.
Syntax
Arg1 Not Begins With Arg2
Where:
Example
Region@DESC Not Begins with "South"
Returns TRUE if region is "Northeast", "North", and so on.
Not Between
Returns TRUE if a specified value does not lie in between two given
boundaries.

Syntax
Value Not Between Boundary1 and Boundary2
Where:
Value, Boundary1, and Boundary2 must be of the same data type. They
can be of any data type that MicroStrategy supports.
Example
Year not between 2000 and 2005
Returns TRUE if Year = 1999
Not Contains
Returns TRUE if a text string does not contain the specified text pattern.
Syntax
Arg1 Not Contains Arg2
Where:
Arg1 and Arg2 must be of Text data type.
Example
Region@DESC Not Contains "South"
Returns TRUE if region is "North", "West", and so on.
Not Ends With

Returns TRUE if a text string does not end with the specified text pattern.
Syntax
Arg1 Not Ends With Arg2
Where:
Arg1 and Arg2 must be of Text data type.

Example
Region@DESC Not Ends With "East"
Returns TRUE if region is "Northwest", "West", and so on.
Not In
Returns TRUE if a given value is not in the specified list of values.
Syntax
Arg1 not in Arg2
Where:
Arg2 must be a list with one or more elements. Arg1 must be of the same
data type as the elements in Arg2. Arg1 and Arg2 can be of any data type
that MicroStrategy supports.
Example
Year not in (2002, 2003)
Returns TRUE if year is 2000, 2001, and so on.
Not Like
Returns TRUE if a text string does not match the specified text pattern;
otherwise, returns FALSE.
Syntax
Arg1 Not Like Arg2
Where:
Example
Region@DESC not like "South%"
Returns TRUE if region is "Northeast", "North", "Mid South", and so on.

Comparison for Rank Operators

These operators compare rank values. The Comparison for rank operators
are:
Less Than or Equal Enhanced (*<=)

Returns TRUE for values in a list which are less than or equal to a specified
condition. The list values are generated from the Rank function.
Syntax
ValueList *<= Condition
Where:
l ValueList is the list of the rank values of a metric.
l Condition is the value to be compared.
Example
Build a Set Qualification filter with
l Metric: M1
l Function: Rank
l Operator: *<=
l Value: 2
Returns TRUE for the values of M1 whose rank values are less than or equal
to 2.
Not Equal Enhanced (*<>)

Returns TRUE for values in a list which are not equal to a specified condition.
The list values are generated from the Rank function.
Syntax
ValueList *<> Condition

Where:
Example
l Metric: M1
l Function: Rank
l Operator: *<>
l Value: 2
Returns TRUE for the values of M1 whose rank values are not equal to 2.
Equal Enhanced (*=)

Returns TRUE for values in a list which are equal to a specified condition.
The list values are generated from the Rank function.
Syntax
ValueList *= Condition
Where:
Example
l Metric: M1
l Function: Rank

l Operator: *=
l Value: 2
Returns TRUE for the values of M1 whose rank values are equal to 2.
Greater Than or Equal Enhanced (*>=)

Returns TRUE for values in a list which are greater than or equal to a
specified condition. The list values are generated from the Rank function.
Syntax
ValueList *>= Condition
Where:
• ValueList is the list of the rank values of a metric.
• Condition is the value to be compared.
Example
l Metric: M1
l Function: Rank
l Operator: *>=
l Value: 2
Return TRUE for the values of M1 whose rank values are greater than or
equal to 2.
Between Enhanced (*Between)

Returns TRUE for the values in a list of values that are in a specified range.
The values are generated from the Rank function.
Syntax
ValueList *Between Condition1 and Condition2

Where:
l Condition1 and Condition2 set the range of values to be compared.
Usage Notes
*Between is inclusive which means any value greater than or equal to

Condition1 and less than or equal to Condition2 will return TRUE.
Example
l Metric: M1
l Function: Rank
l Operator: *between
l Value: 2 and
l Value: 8
Returns TRUE for the values of M1 whose rank values are between 2 and 8.
Not Between Enhanced (Not *Between)

Returns TRUE if a list of rank values is not within a specified range. The
values are generated from the Rank function.
Syntax
ValueList Not *Between Condition1 and Condition2
Where:
l Condition1 and Condition2 set the range of values to be compared.
Usage Notes

Values equal to Condition1 and Condition2 are not satisfied with the
condition.
Example
l Metric: M1
l Function: Rank
l Operator: *Not between
l Value: 2 and
l Value: 8
Return TRUE for the values of M1 whose rank values are less than 2 or
greater than 8.
Logical Operators
The logical operators allow certain conditions to be applied to two sets of
filter expressions simultaneously.
And
Returns TRUE if both the specified conditions are TRUE; otherwise, returns
FALSE.
Syntax
Arg1 And Arg2
Where:
Arg1 and Arg2 are conditional expressions. The condition can contain
metrics, comparison and logical operations, functions, and constants. The
condition must be evaluated as TRUE or FALSE.
Example

(Cost < 1000) And (Freight < 500)
This condition will return TRUE only if both, the Cost is less than 1000, and
the Freight is less than 500.
IF
Returns a value if the specified condition is TRUE; otherwise, a default value
is returned. This is a single value function.
Syntax
IF (Condition, TrueBranch, FalseBranch)

Where:
l Condition is the conditional expression. The condition can contain

metrics, comparison and logical operations, and constants. The condition
must be evaluated to be TRUE or FALSE.
l TrueBranch is a constant or metric value to return if the condition is

TRUE.
l FalseBranch is a constant or metric value to return if the condition is

FALSE.
Usage Notes
FalseBranch must be provided; otherwise the return value is undefined.
Example
IF ((Total Revenue < 300000), 0, 1)

This metric returns 0 if the Total Revenue is less than 300,000; otherwise, it
returns 1.
Not
Returns TRUE if the specified condition is FALSE, and FALSE if the condition
is TRUE.
Syntax
Not(Arg1)
Where:
Arg1 is the conditional expression. The condition can contain metrics,

comparison and logical operations, and constants. The condition must be
evaluated as TRUE or FALSE.
Example
Not ((Profit <= 0))
This condition returns TRUE only if the profit is greater than zero.
Or
Returns FALSE if both the specified conditions are FALSE; else returns
TRUE.
Syntax
Arg1 Or Arg2
Where:
Arg1 and Arg2 are conditional expressions. The condition can contain
metrics, comparison and logical operations, and constants. The condition
must be evaluated as TRUE or FALSE.

Example
(Cost <= 1000) Or (Freight <= 500)
This condition returns FALSE only if the Cost is greater than 1000, and the
Freight is greater than 500.

PLUG -I N PACKAGE
FUN CTION S

The functions in this chapter represent the more advanced functions

available in MicroStrategy, including data mining, financial, mathematical,
and statistical functions. For more information on the plug-in functions and
how to install the Function Plug-in Wizard, see Using Custom Plug-In
Functions, page 79.
Each section briefly describes the category of function and then lists each
function along with information designed to provide data necessary for
understanding and implementing an individual function. The information
provided for each function includes:
l An explanation of the data returned by the function
l The syntax of the function including function name, the available

parameters, the parameter setting defaults, and the types of data possible
for use with the function
l The mathematical expression illustrating exactly how the calculation is

defined in MicroStrategy (if applicable)
l Usage notes describing any error conditions, invalid data types, or key
items to know before using the function (if applicable)
l An example of the function in use; this can be either a report example or a

simple text description of the data returned based on the specified input
For a list of databases and the functions they support, see Appendix A,
MicroStrategy and Database Support for Functions.
The following categories of functions are covered:
Data Mining Functions

Data mining generally refers to examining a large amount of data to extract
valuable information. The data mining process uses predictive models based
on existing and historical data to project potential outcome for business

activities and transactions. MicroStrategy Data Mining Services facilitates

the development and deployment of these predictive models.
Data mining is covered in the Data Mining Services chapter of the Advanced
Reporting Help. The Data Mining Services chapter introduces MicroStrategy
Data Mining Services, which includes these features:
l Using MicroStrategy and the Training Metric Wizard to create multi-

variable regression predictive models
l Support for importing third-party predictive models using the PMML

industry standard
l A Predictive Model Viewer that visualizes the predictive model
l A set of sample predictive metrics and reports incorporated into Customer

Analysis Module (CAM)
In addition, the Data Mining Services chapter of the Advanced Reporting

Help describes the process of how to create and use predictive models with
MicroStrategy and provides a business case for illustration.
The data mining functions that are available within MicroStrategy are
employed when using standard MicroStrategy Data Mining Services
interfaces and techniques, which includes the Training Metric Wizard and
importing third-party predictive models. To ensure proper functionality, it is
recommended to use these MicroStrategy data mining functions within the
Data Mining Services interfaces and techniques, rather than manually
defining the values and parameters for these functions.
Functions for R Integration

MicroStrategy supports the integration and deployment of analytics from the
R statistical environment to MicroStrategy Web or Workstation. Customers
interested in deploying analytics from the R programming language into
MicroStrategy can do so using the R Integration Pack, available on the
MicroStrategy GitHub site.

The third-party R environment is freely available, as a separate download,

from CRAN-R.
Once you have downloaded and configured the R Integration Pack, there are
several functions available in MicroStrategy Web or Workstation that allow
you to deploy your R analytics. These functions include:
For the functions listed below, scalars are variables with a single value
while vectors are variables with one or more values.
l RScript: Supports R scripts that use vectors for the inputs and output.
Sorting is also supported.
l RScriptAgg: Supports R scripts that use vectors for the inputs and a
scalar output. Sorting is also supported.
l RScriptAggU: Supports R scripts that use vectors for the inputs and a
scalar output. This is a version of the RScriptAgg function that does not
include sorting.
l RScriptSimple: Supports R scripts that use scalar for the inputs and
output. Sorting is not supported.
l RScriptU: Supports R scripts that use vectors for the inputs and output.
This is a version of the RScript function that does not include sorting.
The easiest way to create a metric expression that utilizes these functions
A metric expression that can then be used in a derived metric to include the
statistical analysis of an R script in a MicroStrategy Web or Workstation
dossier. The R Integration Pack User Guide provides steps to use the
deployR utility to create metric expressions for R scripts.
When defining these functions in MicroStrategy Web or Workstation, you

have the following options:
l Value1, ..., ValueN: The MicroStrategy metrics that act as inputs for
the R script.

l BooleanParam1, ..., BooleanParam9: A set of boolean parameters

that allow you to pass boolean values into R scalar values that do not
change from execution to execution. If you use any boolean parameters in
your R script, you can define the default value for the parameter when
using the function in MicroStrategy Web or Workstation.
l NumericParam1, ..., NumericParam9: A set of numeric parameters

that allow you to pass numeric values into R scalar values that do not
change from execution to execution. If you use any numeric parameters in
your R script, you can define the default value for the parameter when
using the function in MicroStrategy Web or Workstation.
l StringParam1, ..., StringParam9: A set of string parameters that

allow you to pass string values into R scalar values that do not change
from execution to execution. If you use any string parameters in your R
script, you can define the default value for the parameter when using the
function in MicroStrategy Web or Workstation.
l _WorkingDir: The R scripts working directory, which is used to store

various supporting files.
l _OutputVar: The R variable that is used as the output for the metric. If
there is more than one output, the first output is considered the default
output unless otherwise specified here.
l _NullsAllowed: Controls whether records containing null values are to

be passed in as inputs to your analytic:
l True (default): Null values are included in the analysis.
l False: All records containing null values are eliminated from the
analysis.
l _CheckInputCount: Controls whether MicroStrategy ensures that the

number of inputs to the metric exactly matches exactly the number of
inputs specified in the function's signature:

l True (default): If the number of inputs is different, a warning message is

returned when using the R script in MicroStrategy.
l False: If the number of inputs is different, the script execution will

attempt to proceed.
l _RScriptFile: The directory where the R script file is stored.
l _InputNames: The MicroStrategy metric names that are used as inputs to

the R script. For the R environment to be able to use the MicroStrategy
metric names in objects and graphics that R generates, the names
associated with the inputs from MicroStrategy need to be passed to R.
l _Params: For internal use only.
l BreakBy: The logical level where the calculation of values for an

expression restarts. You can select the break by attribute from the drop-
down list.
l SortBy: Controls the sorting of records before the data is passed to R. To

specify a particular sorting criterion, you can select the sort by value from
the drop-down list. Use the button to the right of the drop-down list to
define whether the sort is ascending or descending.
Financial Functions
The financial functions plug-in package in MicroStrategy provides access to
many standard financial calculations. All finance-related calculations are
performed by the MicroStrategy Analytical Engine, regardless of the
database environment.
Accrint (Accrued Interest)

Accrued interest is interest that has accumulated over a period of time, but
has not yet been paid. This function returns the amount of accrued interest
on a security that pays periodic interest.
Syntax

Accrint <Par = 1000, Basis> (Issue, FirstInterest, Settlement,

Rate, Frequency)
Where:
Issue is the issue date.
FirstInterest is the first date on which interest is accrued.
Settlement is the settlement date. This is the date, after issue, on which
the security is traded.
Rate is the annual coupon rate.
Par is a parameter that indicates the par value. The default, and only valid
value is 1000.
Frequency is the number of coupon payments per year. The valid values
are, 1, 2, and 4 where annual payments =1, semiannual payments =2, and
quarterly payments =4.
Basis is a parameter that indicates the time-count basis to be used. The

default value for Basis is 0, which is typically used by American agencies
and assumes 30-day months and 360-day years (30/360). Possible values
for this parameter are listed in the following table.
Basis value Application
0 (30/360) Assumes 30 days in each month, 360 days in each year.
Assumes actual number of days in each month, actual number of days in

1 (actual/actual)
each year.
2 (actual/360) Assumes actual number of days in each month, 360 days in each year.
Used by European agencies, assumes the same values as “0” for

4 (30/360)
American institutions.
Expression

Where:
Ai is the number of accrued days for the ith quasi-coupon period within an
odd period
NC is the number of quasi-coupon periods that fit an odd period (if this
period contains a fraction, that fraction is rounded up to the nearest integer)
NLi is the normal length, in days, of the ith quasi-coupon period within an
odd period
Usage Notes
If Issue, FirstInterest, Settlement, or Frequency is not an integer, it is

truncated
The engine returns an empty cell if:
l Issue, FirstInterest, or Settlement is not a valid date
l
Par ¹ 1000
l
Issue ³ Settlement
l Frequency is a value other than 1, 2, or 4
l The Issue date, the FirstInterest date and the Settlement date should be
included within single quotations in the expression for the expression to be
considered as a valid expression
Example
This example displays the expression built using Accrint for a treasury bond
with the following terms:

l March 22, 2003, issue date
l June 20, 2003, first interest date
l September 16, 2003, settlement date
l 10.0 percent coupon
l $1,000 par value
l Frequency is semiannual
l Basis is 30/360
The accrued interest is defined as:
Accrint <Par=1000, Basis=0>

(‘3/22/2003’,’9/16/2003’,’6/20/2003’,0.1,2) {~+}
Accrintm (Accrued Interest at Maturity)

Accrued interest is interest that has accumulated over a period of time, but
has not yet been paid. This function returns the accrued interest amount on
a security that pays periodic interest at maturity.
Syntax
Accrintm <Par = 1000, Basis> (Issue, Maturity, Rate)
Where:
Maturity is the maturity date. This is the date on which the coupon
expires.
Par is a constant that indicates the par value. The default, and only valid,
value is 1000.



1 (actual/actual)
each year.

4 (30/360)
Expression
Where:
A is the accrued time (for interest-at-maturity items, the value used is the
number of days from issue to maturity)
D is the annual-yield basis
Usage Notes
If Issue or Maturity is not an integer, it is truncated.
The engine returns an empty cell if
l Issue or Maturity is not a valid date
l Rate £ 0
l Par ¹ 1000

l The Issue date and the Maturity date should be included within single
quotations in the expression for the expression to be considered as a valid
expression.
Example
This example displays the expression built using the Accrintm function for a
note with the following terms:
l March 22, 2001, issue date
l June 20, 2003, maturity date
l 10.0 percent coupon
l $1,000 par value
l Frequency is semiannual
l Basis is Actual/365
The accrued interest at maturity is defined as:
Accrintm <Par=1000, Basis=3> (‘3/22/2001’,’6/20/2001’,0.1){~+}
Coupdaybs (Coupon Period, Beginning to Settlement)

Returns the number of days from the beginning of a coupon period to the
settlement date.
Syntax
Coupdaybs <Basis> (Settlement, Maturity, Frequency)
Where:
Maturity is the maturity date. This is the date on which the security
expires.

are 1,2 and 4 where annual payments =1, semiannual payments =2, and


1 (actual/actual)
each year.

4 (30/360)
Usage Notes
If an argument is not an integer, it is truncated.
The Settlement date and the Maturity date should be included within single
expression.
Coupdays (Coupon Period, Number of Days with Settlement)

Returns the number of days in the coupon period that contains the
settlement date.
Syntax
Coupdays <Basis> (Settlement, Maturity, Frequency)
Where:

expires.
are 1, 2, and 4 where annual payments =1, semiannual payments =2, and


1 (actual/actual)
each year.

4 (30/360)
Usage Notes
The number of days between milestones is computed depending on the

chosen day basis
Coupon functions are defined against the maturity day, depending on

frequency
expression

Coupdaysnc (Coupon Period, Settlement to Next Coupon)

Returns the number of days from the settlement date to the next coupon
date.
Syntax
Coupdaysnc <Basis = 0> (Settlement, Maturity, Frequency)
Where:
expires.


1 (actual/actual)
each year.

4 (30/360)
Usage Notes

l Settlement or Maturity is not a valid date.
l Frequency is a number other than 1, 2, or 4.
l Settlement ³ Maturity.
expression.
Coupncd (Next Date After Settlement)

Returns a number that represents the next coupon date after settlement.
Syntax
Coupncd <Basis> (Settlement, Maturity, Frequency)
Where:
expires.



1 (actual/actual)
each year.

4 (30/360)
Usage Notes
l Settlement or Maturity is not a valid date
l Frequency is a number other than 1, 2, or 4
l
Settlement ³ Maturity
expression.
Coupnum (Coupon, Number Payable Between Settlement and

Maturity)
Returns the number of coupons payable between the settlement date and
maturity date, rounded up to the nearest whole coupon.
Syntax
Coupnum <Basis> (Settlement, Maturity, Frequency)
Where:

expires.
are, 1, 2, and 4 where annual payments =1, semiannual payments =2, and


1 (actual/actual)
each year.

4 (30/360)
Usage Notes
The number of days between milestones is computed depending on the

chosen day basis.
Coupon functions are defined against the maturity day, depending on

frequency.
expression.

Couppcd (Coupon Date, Previous)

Returns a number that represents the coupon date before settlement date.
Syntax
Couppcd <Basis> (Settlement, Maturity, Frequency)
Where:
expires.
Frequency is the number of payments per year. The valid values are 1, 2,
and 4 where annual payments =1, semiannual payments =2, and quarterly
payments =4.


1 (actual/actual)
each year.

4 (30/60)
Usage Notes

l Frequency has a value other than 1, 2, or 4.
l
Settlement ³ Maturity.
expression.
Cumipmt (Cumulative Interest Paid)

Returns the cumulative interest paid on a loan over a specified period of
time.
Syntax
Cumipmt <Type = 0> (Rate, Nper, Pv, Start, End)
Where:
Rate is the interest rate.
Nper is the total number of payment periods.
Pv is the present loan value.
Start is the first period in the calculation. Payment period counting begins
at 1.
End is the last period in the calculation.
Type is the timing of the payment.
Usage Notes
Consistency is important when specifying Rate and Nper:
l For monthly payments on a four-year loan at an annual interest rate of

12%, Rate = 12%/12; Nper = 4 × 12.

l For annual payments on a four-year loan at an annual interest rate of 12%,

Rate = 12%; Nper = 4.
If Nper, Start, End, or Type is not an integer, it is truncated.
l Rate £ 0, Nper £ 0, or Pv £ 0
l Start < 1 or End < 1
l Start > End
l Type ¹ 0
Example
A home mortgage loan has the following terms:
l Interest rate: 9.00 percent per annum (rate = 9.00% ÷ 12 = 0.0075)
l Term: 30 years (nper = 30 × 12 = 360)
l Present value: $125,000
The total interest paid in the second year of payments (periods 13 through
24) is defined as follows:
CUMIPMT(0.0075,360,125000,13,24) equals -11135.23
The interest paid in a single payment in the first month is defined as follows:
CUMIPMT(0.0075,360,125000,1,1) equals -937.50
Cumprinc (Cumulative Principal Paid)

Returns the cumulative principal paid on a loan over a specified period of
time.
Syntax
Cumprinc <Type = 0> (Rate, Nper, Pv, Start, End)
Where:

Pv is the present loan value.
Start is the first period in the calculation. Payment period counting begins
at 1.
End is the last period in the calculation.
Type is a parameter that indicates the timing of the payment.
Usage Notes
Consistency is important when specifying Rate and Nper:
l For monthly payments on a four-year loan at an annual interest rate of

12%, Rate = 12%/12; Nper = 4 × 12.
l For annual payments on a four-year loan at an annual interest rate of 12%,

Rate = 12%; Nper = 4.
If Nper, Start, End, or Type is not an integer, it is truncated.
l Rate £ 0, Nper £ 0, or Pv £ 0
l Start< 1 or End < 1
l Start > End
l Type ¹ 0
Example
A home mortgage loan has the following terms:
l Interest rate: 9.00 percent per annum (rate = 9.00% ÷ 12 = 0.0075)
l Term: 30 years (nper = 30 × 12 = 360)
l Present value: $125,000

The total principal paid in the second year of payments (periods 13 through
24) is defined as follows:
CUMPRINC(0.0075,360,125000,13,24) equals -934.1071
The principal paid in a single payment in the first month is defined as

follows:
CUMPRINC(0.0075,360,125000,1,1) equals -68.27827
Db (Fixed-Declining Balance (Asset Depreciation))

Returns the depreciation of an asset for a specified period, using the fixed-
declining-balance method. This is a method of accelerated depreciation
which is faster than straight line depreciation early in the life of the asset.
Syntax
Db <Month> (Cost, Salvage, Life, Period)
Where:
Cost is the initial cost of the asset.
Salvage is the value of the asset at the end of the depreciation period.
Life is the number of periods over which the asset is depreciated. This is
sometimes referred to as the useful life of an asset.
Period is the period of time over which depreciation is calculated.
Month is an integer indicating the number of months in the first year.
Expression
To calculate depreciation over any given period:
Where:

To calculate depreciation for the first period:
To calculate depreciation for the last period:
Ddb (Double-Declining Balance (Asset Depreciation))

Returns the depreciation of an asset for a specified period, using the
double-declining-balance method. Double declining balance is a method of
accelerated depreciation that is twice as fast as the straight line
depreciation method.
Syntax
Ddb <Factor> (Cost, Salvage, Life, Period)
Where:
Cost is the initial cost.
Salvage is the cost at the end of depreciation.
Life is the number of periods over which depreciation occurs.
Period is the period for which depreciation is calculated.
Factor is the rate at which the balance declines.
Expression
Usage Notes
All arguments must be positive values.
The double-declining balance method computes depreciation at an

accelerated rate: depreciation is highest during the first period and
decreases with each succeeding period.

Disc (Discount Rate for a Security)

Returns the discount rate on a security. The discount rate is used to
calculate the present value of expected or future benefits and costs.
Syntax
Disc <Basis> (Settlement, Maturity, Price, Redemption)
Where:
expires.
Price is the price per $100 of face value.
Redemption is the redemption value per $100 of face value.


1 (actual/actual)
each year.

4 (30/60)
Expression

Where:
B is the number of days in a year (see Basis)
DSM is the number of days between settlement and maturity
Usage Notes
If Settlement or Maturity is not an integer, it is truncated.
l The engine returns an empty cell if
l Price £ 0 or Redemption £ 0.
expression.
Dollarde (Dollar Price, Converted from Fraction to Decimal)

Converts a dollar price expressed as a fraction into a dollar price expressed
as a decimal number.
Syntax
Dollarde (FractionalDollar, Decimal)
Expression
Where:
price is the dollar value to be converted

fraction is the fraction denominator value as an integer
Dollarfr (Dollar Price, Converted from Decimal to Fraction)

Converts a dollar price expressed as a decimal into a dollar price expressed
as a fraction.
Syntax
Dollarfr (DecimalDollar, Fraction)
Where:
DecimalDollar is the dollar value to be converted
Fraction is the fraction denominator value as an integer
Duration
Returns the Macauley duration for an assumed par value of $100. Duration
is the weighted average of the present value of the cash flows and is used as
a measure of a bond price’s response to changes in yield. Bonds with higher
durations face higher risk from changes in interest rates.
Syntax
Duration <Basis> (Settlement, Maturity, CouponRate, YieldRate,

Frequency)
Where:
expires.
CouponRate is the annual interest rate of the coupon.
YieldRate is the annual yield.



1 (actual/actual)
each year.

4 (30/60)
Expression
Where:
DSC is the number of days from settlement to next coupon date
E is the number of days in settlement’s coupon period

N is the number of coupons from settlement to maturity
Usage Notes
expression.
Effect (Effective Annual Interest Rate)

Returns the effective annual interest rate on a given amount. This is the
actual annual rate or yield on a loan.
Syntax
Effect(NominalRate, Npery)
Where:
NominalRate is the nominal interest rate.
Npery is the number of compounding periods per year.
Expression
Usage Notes
If Npery is not an integer, it is truncated.
l Either argument is nonnumeric
l NominalRate £ 0
l Npery <= 1.

Fv (Future Value)
Returns the future value of an investment based on periodic, constant
payments and a constant interest rate. The future value is equal to the
present value plus the amount of accumulated interest that would be earned.
This can be applied to future payments, with interest calculated for each
payment.
Syntax
Fv <Type> (Rate, Nper, Pmt, Pv)
Where:
Rate is the interest rate per period.
Pmt is the (fixed) amount paid each period. This typically includes principal
and interest, but no other fees or taxes.
Pv is the present value (lump-sum amount) of future payments. If no value is

provided, it is assumed to be 0 (zero).
Type is a parameter that indicates when payments are due.
Usage Notes
In all cases, disbursements, such as deposits, are represented by negative

numbers; funds received, such as dividends, are represented by positive
numbers.
Use the Fvschedule function for payments made with a variable or

adjustable interest rate.
Fvschedule (Future Value Schedule)

Returns the future value of an initial principal after a series of compound-
interest rates are applied. Use this function to calculate the future value of
an investment that has a variable or adjustable rate.
Syntax

Fvschedule <Basis> (Argument, PV)
Where:
Argument is a set of applicable interest rates.
PV is the present value of the investment.


1 (actual/actual)
each year.

4 (30/60)
Usage Notes
If Argument is nonnumeric, the engine returns an empty cell.
Use the Fv function for payments made with a constant interest rate.
Intrate (Interest Rate)

Returns the interest rate for a fully invested security. The interest rate is the
percentage value that is multiplied by the principal to determine the amount
of interest due or paid in a given period of time.
Syntax
Intrate <Basis> (Settlement, Maturity, Investment, Redemption)

Where:
expires.
Investment is the amount invested in the security.
Redemption is the amount actually received for the security.


1 (actual/actual)
each year.

4 (30/60)
Expression
Where:
Redemption is the amount actually received for the security

Investment is the amount invested in the security
B is the number of days in a year, depending on year basis
DIM is the number of days from settlement to maturity
Usage Notes
expression.
Ipmt (Interest Payment)

Returns the interest payment for a given period, based on periodic, constant
payments and a constant interest rate.
Syntax
Ipmt < FV, Type> (Rate, Period, Nperiod, PV)
Where:
Rate is the interest rate per period.
Period is the period for which the interest payment is sought. The valid
range is from 1 to Nperiod.
Nperiod is the total number of payment periods.
PV is the present value. This is the total value that a series of future
payments has today.
FV is a parameter that indicates the future value, also called the cash
balance, expected after the last payment is made.
Type is a parameter that indicates when payments are due.
Usage Notes
For this function, consistency in the units used is necessary:

l Assuming monthly payments on a four-year loan at 12% annual interest,

Nperiod should be 4 × 12.
l Assuming annual payments on a four-year loan at 12% annual interest,

Nperiod should be 4.
For this function, disbursements (such as deposits to savings) are

represented by negative numbers; funds received (such as dividend checks)
are represented by positive numbers.
IRR (Internal Rate of Return)

Returns the internal rate of return on a set of payments. The internal rate of
return is the interest rate received for an investment consisting of payments
that occur periodically. These payments do not need to be equal in value,
but they must occur at regular intervals, such as monthly, quarterly, or
annually.
To calculate the internal rate of return for payments that do not occur at
regular intervals, see XIRR (Internal Rate of Return for Payments at
Irregular Intervals), page 405.
Syntax
IRR <FactID, Guess, METHOD, Precision, SortBy> (Argument)
Where:
FactID is a parameter that forces a calculation to take place on a fact table

that contains the selected fact.
Guess is a parameter that indicates an estimate assumed to be close to the

result sought. The default value is 0.1, which can be thought of as
representing an expected internal rate of return of 10 percent.
Method is a parameter that determines the algorithm used to calculate the

IRR function. You can use the following values:

l 0 (default): The IRR function uses the secant method to determine the
internal rate of return. This is the recommended method as it can adjust
the initial guess as required to return a result.
l 1: The IRR function uses Newton’s method to determine the internal rate
of return. This method can return results similar to those returned for
internal rate of return functions that are supplied with Microsoft Excel.
Precision is a parameter that determines how close the evaluation of a

potential solution must be to zero to be considered valid.
Argument is a fact or metric representing a list of numbers for which the

internal rate of return is sought.
SortBy is a parameter that defines the order of calculation. For more

information see BreakBy and SortBy Parameters, page 27.
Usage Notes
The list of values must contain at least one positive entry and one negative
entry.
Payments, whether positive or negative, must be entered in the desired

sequence, as this is the order in which entries are interpreted for
calculation.
Mduration (Modified Duration)

Returns the modified Macauley duration, which is the weighted average of
present payments, for a security with an assumed par value of $100. This
function is used as a measure of bond-price response to variations in yield.
Syntax
Mduration <Basis> (Settlement, Maturity, CouponRate, YieldRate,

Frequency)
Where:

Maturity is the date of maturity. This is the date on which the security
expires.
Basis is a parameter that indicates the time-count basis used. The default
value for Basis is 0, which is typically used by American agencies and
assumes 30-day months and 360-day years (30/360). Possible values for
this parameter are listed in the following table.

1 (actual/actual)
each year.

4 (30/60)
Expression
Where:

yld is the security annual yield
frequency is the number of coupon payments per year
Usage Notes
If Settlement, Maturity, or Frequency is not an integer, it is truncated.
l Either Settlement or Maturity is not a valid date
l CouponRate < 0
l YieldRate < 0
l Frequency is a number other than 1, 2, or 4
l Settlement ³ Maturity
expression.
MIRR (Modified Internal Rate of Return)

Returns the modified internal rate of return on a set of periodic payments.
This function takes into consideration both the cost of investment and the
interest received on investment.
To calculate the internal rate of return for payments that do not occur at
regular intervals, see XIRR (Internal Rate of Return for Payments at
Irregular Intervals), page 405.
Syntax
MIRR <FactID> (Argument, FinancialRate, ReinvestRate)
Where:


Argument is a list of values containing a set of numbers. Entries in this list

can represent payments (negative values) or income (positive values)
occurring at regular intervals.
FinancialRate is the interest rate paid.
ReinvestRate is the interest rate received on reinvested funds.
Expression
Usage Notes
If Argument does not contain at least one positive entry (income) and one
negative entry (payments), the engine returns an empty cell.
In calculating the return for this function, the order in which entries are
provided is assumed to be the order in which funds flow; it is important,
therefore, that payments and income be entered in the correct sequence.
Nominal (Nominal Annual Interest Rate)

Returns the nominal annual interest rate on an investment, based on the
effective rate and the number of compounding periods per year. This is the
interest rate that is not adjusted for actual or expected inflation.
Syntax
Nominal(EffectiveRate, Npery)
Where:
EffectiveRate is the effective interest rate.
Npery is the number of compounding periods per year.
Expression

Usage Notes
If Npery is not an integer, it is truncated.
l EffectiveRate £ 0
l Npery < 1
Nper (Number of Investment Periods)

Returns the number of periods for an investment, based on periodic,
constant payments and a constant interest rate.
Syntax
Nper <Type> (Rate, Pmt, PV, FV)
Where:
Rate is the internal rate for each period.
Pmt is the amount paid each period. Typically it includes principal and
interest, but no other fees or taxes.
PV is the present value of a series of future payments. If this value is

omitted, is assumed to be 0 (zero).
FV is the value that remains after the last payment is made.
Type is a number that indicates when payments are due.
NPV (Net Present Value of an Investment)

Returns the net present value of an investment based on a discount rate and
a set of future payments (negative values) and income (positive values).

All payments and incomes must be equally spaced in time, and occur at the
end of each period. To calculate the net present value of an investment in
which payments and income do not occur at regular intervals, see XNPV
(Net Present Value of an Investment for Payments or Incomes at Irregular
Intervals), page 408.
Syntax
NPV <FactID, SortBy> (Argument, Rate)
Where:


information see BreakBy and SortBy Parameters, page 27.
Argument is a fact or metric representing a list of values containing either

payment or income figures.
Rate is the discount rate for the length of a period.
Expression
Where:
n is the number of entries in Values.
Usage Notes
All entries in Values must be equally spaced in time, and occur at the end of
each period.

The order of the entries in Values is assumed to be the order in which funds
flow; payment and income entries must, therefore, be provided in the correct
sequence.
Investment begins one period before the date on which the first entry in
Values occurs, and ends with the last entry in that array.
If the first entry in Values occurs at the beginning of a period, it must be

added to the result of the function.
The primary differences between this function and the Present Value (PV)
function are that
l PV allows fund flow to occur either at the beginning or at the end of a

period.
l PV fund-flow occurrences must be constant across the life of the

investment.
Oddfprice (Odd-First-Period Price)

Returns the price per $100 face value of a security having an odd (short or
long) first period.
Syntax
Oddfprice <Basis> (Settlement, Maturity, Issue, FirstCoupon,

CouponRate, YieldRate, Redemption, Frequency)
Where:
expires.
FirstCoupon is the first-coupon date.

Basis is a parameter that indicates the time-count basis used. The default
value for Basis is 0, which is typically used by American agencies and
assumes 30-day months and 360-day years (30/360). Possible values for
this parameter are listed in the following table.

1 (actual/actual)
each year.

4 (30/60)
Expression
There are two expressions for this function:
l Odd Short First Coupon: for securities with a short first period
l Odd Long First Coupon: for securities with a long first period
Odd Short First Coupon

Where:
A is the number of days from beginning of coupon period to settlement date

(accrued days)
DFC is the number of days from the beginning of odd first coupon to first
coupon date
E is the number of days in coupon period
N is the number of coupons payable between settlement date and

redemption date; if this number contains a fraction, it is raised to the next
whole number
Odd Long First Coupon
Where:
Ai is the number of days from beginning of the ith quasi-coupon period

within odd period
DCi is the Number of days from date to first quasi-coupon (i=1) or number
of days in quasi-coupons (i=2,..., i=NC)
DSC is the Number of days from settlement to next coupon date
E is the Number of days in coupon period
N is the Number of coupons payable between the first real coupon date and
redemption date; if this number contains a fraction, it is raised to the next
whole number
NC is the Number of quasi-coupon periods that fit in odd period; if this

number contains a fraction it is raised to the next whole number

NLi is the Normal length in days of the full ith quasi-coupon period within
odd period
Nq is the Number of whole quasi-coupon periods between settlement date

and first coupon
Usage Notes
expression.
Oddfyield (Odd-First-Period Yield)

Returns the yield of a security that has an odd (short or long) first period.
Syntax
Oddfyield <Basis> (Settlement, Maturity, Issue, FirstCoupon,

CouponRate, Price, Redemption, Frequency)
Where:
expires.
FirstCoupon is the first coupon date.
Price is the price.
payments =4.



1 (actual/actual)
each year.

4 (30/60)
Usage Notes
If Settlement, Maturity, Issue, or FirstCoupon is not an integer, it is

truncated.
l Settlement, Maturity, Issue, or FirstCoupon is not a valid date
l CouponRate < 0
l Price £ 0
l The following date-related condition is not satisfied: Maturity >

FirstCoupon > Settlement > Issue
expression.

Oddlprice (Odd-Last-Period Price)

Returns the price per $100 face value of a security having an odd (short or
long) last period.
Syntax
Oddlprice <Basis> (Settlement, Maturity, LastInterest,

CouponRate, YieldRate, Redemption, Frequency)
Where:
expires.
LastInterest is the last date on which interest is accrued.
payments =4.

1 (actual/actual) Assumes actual number of days in each month, actual number of days in

each year.

4 (30/60)
Expression
Where:
Ai is the Number of accrued days for the ith quasi-coupon period within odd
period counting forward from last interest date before redemption
DCi is the Number of days counted in each ith quasi-coupon period as

delimited by the length of the actual coupon period
NC is the Number of quasi-coupon periods that fit in odd period; if this

number contains a fraction, it is raised to the next whole number
NLi is the Normal length in days of the ith quasi-coupon period within odd
coupon period
DSCi is the Number of days from settlement (or beginning of quasi-coupon

period) to next quasi coupon within odd period (or to maturity date) for each
ith quasi-coupon period.
Usage Notes

expression.
Oddlyield (Odd-Last-Period Yield)

Returns the yield of a security that has an odd (short or long) last period.
Syntax
Oddlyield <Basis> (Settlement, Maturity, LastInterest,

CouponRate, Price, Redemption, Frequency)
Where:
expires.
LastInterest is the security’s last coupon date.
CouponRate is the interest rate.
Price is the price.
payments =4.



1 (actual/actual)
each year.

4 (30/60)
Expression
Where:
Ai is the number of accrued days for the ith, or last, quasi-coupon period
within odd period counting forward from last interest date before redemption.
DCi is the number of days counted in the ith, or last, quasi-coupon period as
delimited by the length of the actual coupon period.
NC is the number of quasi-coupon periods that fit in odd period; if this

number contains a fraction, it is raised to the next whole number.
NLi is the normal length in days of the ith, or last, quasi-coupon period
within odd coupon period.
Usage Notes
The date arguments should be included within single quotations for the
expression to be considered as a valid expression.

Pmt (Payment)
Returns the payment cost on a loan, based on constant payments and a
constant interest rate.
Syntax
Pmt <FV, Type> (Rate, Nper, PV)
Where:
FV a parameter that indicates the future value of the loan. It is the balance to
be attained once all payments are made.
Type is parameter that indicates when payment is due.
Nper is the number of payments.
PV is the present value of the loan, also referred to as principal. This is the
current value of a set of future payments.
Usage Notes
Values returned by this function include principal and interest; they do not
include taxes, reserve payments, or ancillary fees.
Units must be used consistently when calculating payment. For example:
l To calculate monthly payments on a four-year loan at 12% annual interest,

Rate should be 12%/12, Nper should be 4 × 12.
l To calculate annual payments on the same loan, Rate should be 12%,

Nper should be 4.
This function can also be used to determine payments on annuities other

than loans.
Ppmt (Principal Payment)

Returns the payment on the principal for an investment for one given period.

Syntax
Ppmt <Type = 0, FV = 0> (Rate, Per, Nper, PV)
Where:
Type is a parameter that indicates when payment is due.
Per is the period desired.
PV is the present value of the loan.
FV is the future value of the loan.
Expression
Where:
i is the period desired
Price (Price Per $100 Face Value)

Returns the price, per $100 face value, on a security that pays periodic
interest.
Syntax
Price <Basis> (Settlement, Maturity, CouponRate, YieldRate,

Redemption, Frequency)
Where:
expires.

CouponRate is the annual interest rate.
payments =4


1 (actual/actual)
each year.

4 (30/60)
Expression
Where:

E is the number of days in the coupon period in which the settlement date
falls
N is the number of coupons payable between the settlement date and the
redemption date
A is the number of days from the beginning of the coupon period to

settlement date
Usage Notes
If Settlement, Maturity, or Frequency is not an integer, it is truncated.
l CouponRate < 0 or YieldRate < 0
l Redemption £ 0
l Frequency has a value other than 1, 2, or 4
expression.
Pricedisc (Price, Discounted)

Returns the price, per $100 face value, on a discounted security.
Syntax
Pricedisc <Basis> (Settlement, Maturity, DiscRate, Redemption)
Where:

expires.
DiscRate is the discount rate.


1 (actual/actual)
each year.

4 (30/60)
Expression
Where:
DSM is the number of days from settlement to maturity
Usage Notes

l
DiscRate £ 0 or Redemption £ 0
l
Settlement ³ Maturity
expression.
Pricemat (Price at Maturity)

Returns the price, per $100 of face value, on a security that pays interest at
maturity.
Syntax
Pricemat <Basis> (Settlement, Maturity, Issue, CouponRate,

YieldRate)
Where:
expires.
Issue is the issue date. It is expressed as a serial date.
CouponRate is the interest rate on the date of issue.



1 (actual/actual)
each year.

4 (30/60)
Expression
Where:
DSM is the number of days from settlement to maturity
DIM is the number of days from issue to maturity
A is the number of days from issue to settlement
Usage Notes
If Settlement, Maturity, or Issue is not an integer, it is truncated.
l Settlement, Maturity, or Issue is not a valid date
l Rate < 0 or Yield < 0

The Settlement date, the Maturity date, and the Issue date should be
included within single quotations in the expression for the expression to be
considered as a valid expression.
Pv (Present Value)
Returns the present value of an investment.
Syntax
Pv <Type> (Rate, Nper, Pmt, FV)
Where:
Type is a value that indicates when payment is due.
Pmt is the periodic payment for an annuity.
FV is the future value of the loan.
Expression
If rate = 0
Else
Rate (Interest Rate Per Period)

Returns the interest rate per period on a given annuity.
Syntax
Rate <FV, Type, Guess> (Nperiod, Payment, PV)
Where:
l FVis the future value (also called cash balance) expected after the last
payment.
l Typeindicates when payments are due.
l Guess is an estimate assumed to be close to the result sought.
l Nperiod is the total number of payment periods.
l Payment is the payment made for each period. Cannot change over the
life of the annuity. Typically, includes principal and interest, but no other
fees or taxes.
l PV is the present value of the annuity. It is the total amount that a series of
future payments is worth today.
Usage Notes
l For this function, consistency in the units used is necessary:
l Assuming monthly payments on a four-year loan at 12% annual interest,

Nperiod should be 4 × 12.
l Assuming annual payments on a four-year loan at 12% annual interest,

Nperiod should be 4.
Received (Amount Received at Maturity)

Returns the amount received at maturity on a fully invested security.
Syntax
Received <Basis> (Settlement, Maturity, Investment, Discount)
Where:
expires.
Investment is the amount invested.

Discount is the discount rate.


1 (actual/actual)
each year.

4 (30/60)
Expression
Where:
DIM is the number of days between settlement and maturity
Usage Notes

l
Investment £ 0.
l
Discount £ 0.
l
Maturity £ Settlement.
l The Settlement date and the Maturity date should be included within single
expression.
Sln (Straight-Line Depreciation)

Returns the straight-line depreciation of an asset over a given period. This is
the simplest method of depreciation and represents an equal amount
charged for each period during the useful life of the asset after allowing for a
salvage value.
Syntax
Sln(Cost, Salvage, Life)
Where:
Salvage, also called the salvaged value of an asset, is the residual value at
the end of the estimated life span.
Life, also called the useful life of an asset, is the number of periods over
which the asset is being depreciated.
Syd (Sum of Year's Digits Depreciation)

Returns the sum of year’s digits depreciation of an asset over a specified
period. This method of accelerated depreciation results in higher
depreciation early in the life of an asset. It calculates the depreciation for
each year on an inverted scale in proportion to the total of all the digits for
the asset’s useful life.
Syntax

Syd < > (Cost, Salvage, Life, Period)
Where:
Cost is the initial cost.
Salvage, also called the salvaged value of an asset, is the residual value at
the end of the estimated life span.
Life, also called the useful life of an asset, is the number of periods over
which the asset is being depreciated.
Period is the depreciation period. This must have the same number of units
as Life.
Expression
Tbilleq (T-Bill Equity)

Returns the bond-equivalent yield of a Treasury bill.
Syntax
Tbilleq(Settlement, Maturity, Discount)
Where:
Settlement is the settlement date. This is the date on which the security is
traded.
Maturity is the maturity date. This is the date on which the treasury bill
expires.
Expression

Where:
DSM is the number of days between settlement and maturity on 360-day

basis.
Usage Notes
expression.
Tbillprice (T-Bill Price)

Returns the price of a treasury bill per $100 of face value.
Syntax
Tbillprice(Settlement, Maturity, Discount)
Where:
Settlement is the settlement date. This is the date on which the treasury
bill is purchased.
expires.
Expression
Where:
DSM is the number of days from settlement to maturity.
This excludes any maturity date that is more than one calendar year past
Settlement.
Usage Notes

l Discount £ 0
l Settlement > Maturity
l Maturity is more than one year after Settlement
expression.
Tbillyield (T-Bill Yield)

Returns the yield of a treasury bill.
Syntax
Tbillyield < > (Settlement, Maturity, Price)
Where:
Settlement is the settlement date. This is the date on which the treasury
bill is purchased.
expires.
Price is the price per $100 of face value
Expression
Where:
DSM is the number of days from settlement to maturity.

This excludes any maturity date that is more than one calendar year past
Settlement.
Usage Notes
l
Price £ 0
l Settlement > Maturity
l Maturity is more than one year after Settlement
expression.
Vdb (Variable Declining Balance)

Returns the depreciation of an asset over a specified period, including
partial periods.
Syntax
Vdb <Factor> (Cost, Salvage, Life, Period)
Where:
Factor is the rate at which balance declines.
Salvage is the value at the end of the depreciation period.
Life is the number of periods over which depreciation is calculated.
Period is the first period for which depreciation is calculated. This must use
the same number of units as Life.
Usage Notes

All values must be positive.
XIRR (Internal Rate of Return for Payments at Irregular

Intervals)
Returns the internal rate of return on a set of payments that do not occur at
regular intervals. The internal rate of return is the interest rate received for
an investment consisting of payments. These payments do not need to be
equal in value, and do not need to occur at regular intervals.
To calculate the internal rate of return for payments that occur at regular
intervals, see:
l IRR (internal rate of return): Internal rate of return for periodic payments.
l MIRR (modified internal rate of return): Modified internal rate of return for
periodic payments, which takes into consideration both the cost of
investment and the interest received on investment.
Syntax
XIRR <FactID, Guess, SortBy> (Payment, Dates)
Where:

Guess is a parameter that indicates an estimate assumed to be close to the

result sought. The default value is 0.1, which can be thought of as
representing an expected internal rate of return of 10 percent.
Payment is a fact or metric representing a list of numbers for which the

internal rate of return is sought.
Dates is a fact or metric representing a list of dates for the payments that
are included. Ensure that these values use a date/time format.
Usage Notes

The list of values must contain at least one positive entry and one negative
entry.
The XIRR function uses the secant method to determine the internal rate of
return, which adjusts the initial guess as required to return a result. The final
result must be within 0.000001 percent to be considered valid.
Payments, whether positive or negative, must be entered in the desired

sequence, as this is the order in which entries are interpreted for
calculation.
Example
You have an investment where payments are made at irregular intervals.
This payment and payment date information is included in MicroStrategy and
displayed on the simple report shown below:
Since the dates are not at regular intervals such as monthly or quarterly, you
cannot use the IRR function to determine the internal rate of return. Instead,
you can use the XIRR function, which also evaluates the date of each
payment to determine an internal rate of return. A new metric is added to the
report which uses the following definition:
XIRR(Payment, Date){}
The new report now includes this information, as shown below:

The internal rate of return considering the payments and the dates of those
payments is 21.93. Notice that while there is a single value for XIRR, it is
displayed on every row. One way to simplify the display of this information is
to include this data on a dashboard. You can include the XIRR result in a
text field along with additional visualizations of the data, as shown below.

XNPV (Net Present Value of an Investment for Payments or

Incomes at Irregular Intervals)
Returns the net present value of an investment based on a discount rate and
a set of future payments (negative values) and income (positive values).
These payments or incomes do not need to occur at regular intervals. To

calculate the net present value of an investment for payments and incomes
that occur at regular intervals, see NPV (net present value of an
investment).
Syntax
XNPV <FactID> (Payments, Dates, Rate)
Where:

Payments is a fact or metric representing a list of values containing either

payment or income figures.
Dates is a fact or metric representing a list of dates for the payments or

incomes that are included. Ensure that these values use a date/time format.
Rate is the discount rate applied to the cash flows. This is commonly
included in the expression as a constant value.
Expression
Where:
n is the number of entries in payments.

d i is the i th payment date and d 1 is the first payment date.
Usage Notes
The order of the entries in Payments is assumed to be the order in which

funds flow; payment and income entries must, therefore, be provided in the
correct sequence.
Example
You have an investment where payments are made at irregular intervals.
This payment and payment date information is included in MicroStrategy and
displayed on the simple report shown below:
Since the dates are not at regular intervals such as monthly or quarterly, you
cannot use the NPV function to determine the net present value. Instead, you
can use the XNPV function, which also evaluates the date of each payment to
determine a net present value. A new metric is added to the report, which
uses the following definition that includes a rate of 20%:
XNPV(Payment, Date, 0.2){}
The new report now includes this information, as shown below:

The net present value considering the payments, the dates of those
payments, and a rate of 20% is $20,379. Notice that while there is a single
value for XNPV, it is displayed on every row. One way to simplify the display
of this information is to include this data on a dashboard. You can include
the XNPV result in a text field along with additional visualizations of the data,
as shown below.
Yield
Returns the yield on a security that pays periodic interest.

Syntax
Yield <Basis> (Settlement, Maturity, CouponRate, Price,

Redemption, Frequency)
Where:
l Settlement is the settlement date. This is the date on which the security
is purchased.
l Maturity is the maturity date. This is the date on which the security
expires.
l CouponRate is the annual interest rate.
l Price is the price per $100 of face value.
l Redemption is the redemption value per $100 of face value.
l Frequency is the number of payments per year. The valid values are 1, 2,
payments =4.
l Basis is a parameter that indicates the time-count basis to be used. The

and assumes 30-day months and 360-day years (30/360). Other possible
values for this parameter are as follows.

0 (30/ 360) Assu m es 30 d ays i n each m o n t h , 360 d ays i n each year .
1 Assu m es act u al n u m b er o f d ays i n each m o n t h , act u al n u m b er o f d ays
(act u al / act u al ) i n each year .
2 (act u al / 360) Assu m es act u al n u m b er o f d ays i n each m o n t h , 360 d ays i n each year .
3 (act u al / 365) Assu m es act u al n u m b er o f d ays i n each m o n t h , 365 d ays i n each year .
Used b y Eu r o p ean agen ci es, assu m es t h e sam e val u es as “0” f o r
4 (30/ 60)
Am er i can i n st i t u t i o n s.
Expression

• If there is one period (or less) until Redemption, this function is

calculated as follows.
Where:
▫ A is the number of days from the beginning of the period to Settlement
▫ DSR is the number of days from Settlement to Redemption
▫ E is the number of days in a period
• If there is more than one period until Redemption, this function is

calculated through iteration, and is based on the calculation for the
PRICE function. See also Price (price per $100 face value).
Usage notes
• If Settlement, Maturity, or Frequency is not an integer, it is truncated.
• The engine returns an empty cell if:
▫ Either Settlement or Maturity is not a valid date.
▫ Price £ 0 or Redemption £ 0.
▫ Frequency is a number other than 1, 2, or 4.
▫ Settlement > Maturity.
• The Settlement date and the Maturity date should be included within
single quotations in the expression for the expression to be considered
as a valid expression.
Yielddisc (Yield on a Discounted Security)

Returns the annual yield on a discounted security.
Syntax

Yielddisc <Basis> (Settlement, Maturity, Price, Redemption)
Where:
expires.

Basis Value Application
Assumes 30 days in each month; 360 days in

0 (30/360)
each year.
Assumes actual number of days in each month,

1 (actual/actual)
actual number of days in each year.

2 (actual/360)
360 days in each year.

3 (actual/365)
Used by European agencies, assumes the same

4 (30/60)
values as 0 for American institutions.
Usage Notes

l Either Settlement or Maturity is not a valid date.
l
Price £ 0 or Redemption £ 0.
l
Maturity £ Settlement.
expression.
Yieldmat (Yield at Maturity)

Returns the annual yield on a security that pays interest at maturity.
Syntax
Yieldmat < Basis> (Settlement, Maturity, Issue, Rate, Price)
Where:
expires.

0 (30/360) Assumes 30 days in each month; 360 days in

each year.

1 (actual/actual)
actual number of days in each year.

2 (actual/360)

3 (actual/365)
Used by European agencies, assumes the same

4 (30/60)
values as 0 for American institutions.
Usage Notes
If Settlement, Maturity, or Issue is not an integer, it is truncated.
l Rate < 0.
l Price £ 0.
expression.
Mathematical Functions
The category of Mathematical functions contains more complex math
functions than the simple operators found in the Basic Functions. This
category includes exponential, logarithmic, and trigonometric functions.
These functions are calculated by either the MicroStrategy Analytical Engine

or the database. Those not supported by the database are automatically

computed by the Analytical Engine.
Abs (Absolute Value)

Returns the absolute value of a given number. The absolute value of a
number is the number without a sign. This is a single-value function.
Syntax
Abs(Argument)
Where:
Argument is a metric representing a list of real numbers.
Example
This simple example show the results of executing the Abs function on a list
of values. Assuming you have a metric that represents deposit and
withdrawal amounts, the function operates as follows:
Abs(Deposit_Withdrawal) where Deposit_Withdrawal is the

following list of values: +756.25, -315, +145, -980.35, -75, +1250.78, +560.
The resulting value list is 756.25, 315, 145, 980.35, 75, 1250.78, 560. For
another example of using the Abs function, see Example 1: Transformed
Fact, page 12.
Acos (Arc Cosine)

Returns the arc cosine of a given number as an angle. This angle is
calculated in radians and falls between 0 and p . To convert radians to
degrees, multiply by 180/ p . This is a single-value function.
Syntax
Acos(Argument)
Where:

Argument is a metric representing a list of real numbers. The values are the
values of the cosine for the angles sought. This value must be between -1
and 1.
Acosh (Arc Cosine, Hyperbolic)

Acosh returns the inverse hyperbolic cosine of a number. This function is
the inverse of Cosh. Acosh is a single-value function.
For more information on the hyperbolic cosine function, see Cosh (Cosine,
Hyperbolic), page 422.
Syntax
Acosh(Argument)
Where:
Argument is a metric representing a list of real numbers with values greater

than or equal to 1.
Asin (Arc Sine)

Returns the arc sine of a given number as an angle. This angle is calculated
in radians and falls between - p /2 and p /2. To convert radians to degrees,
multiply by 180/ p . This is a single-value function.
Syntax
Asin(Argument)
Where:
values of the sine of the angles sought. These values must be between -1
and 1.
Asinh (Arc Sine, Hyperbolic)

Returns the inverse hyperbolic sine of a number. Asinh is the inverse of the
Sinh function. This is a single-value function.

For more information on the hyperbolic sine function, see Sinh (Sine,
Syntax
Asinh(Argument)
Where:
Atan (Arc Tangent)

Returns the arc tangent of a given number as an angle. This angle is
specified in radians and falls between p /2 and- p/ 2. To convert from radians
to degrees, multiply by 180/ p.
Syntax
Atan(Number)
Where:
Number is the tangent of the angle you want.
Atan2 (Arc Tangent 2)

Returns the arc tangent of the specified x- and y-coordinates. The arc
tangent is the angle from the x-axis to a line containing the origin (0, 0) and
a point with coordinates (x_num, y_num). The angle is given in radians
between - p and p . To convert the result from radians to degrees, multiply by
180/ p . This is a single-value function.
Syntax
Atan2(x_num, y_num)
Where:
x_num is the x-coordinate of the point.
y_num is the y-coordinate of the point.

Atanh (Arc Tangent, Hyperbolic)

Returns the inverse hyperbolic tangent of a given number. Atanh is the
inverse of the Tanh function. This is a single-value function.
For more information on the hyperbolic tangent function, see Tanh (Tangent,
Syntax
Atanh(Number)
Where:
Number is any real number between 1 and -1.
Ceiling (Ceiling Value)

Returns the closest integer greater than or equal to a given number.
Ceiling is often used as a rounding function. This is a single-value
function.
Syntax
Ceiling(Argument)
Where:
Example
Use ceiling calculations in situations where results are expressed as

specified units of measurement. For example, to avoid returns at the cent
level, Ceiling could be used on a product price to round up the figures to
the nearest dollar.
This simple example illustrates how the Ceiling function acts on positive
and negative numbers, and how the results compare to the use of the Floor
function.

Argument values Floor Ceiling
947.67 947.00 948.00
463.33 463.00 464.00
-947.67 -948.00 -947.00
-463.33 -464.00 -463.00
Combine (Combination)
The number of possible combinations for a given number of items selected
as a group from a set.
Syntax
Combine(Number, Number_Chosen)
Where:
Number is the number of items in a set.
Number_Chosen is the number of items to be included in each combination.
Expression
In the expression that follows, Number = n; Number_Chosen = k,
Where:
Usage Notes

In combination calculations, the internal order of the items selected does not
affect the return. This differs from the case of a permutation, in which the
order of the items selected does affect the return. Permutation is a statistical
function.
This function returns an error value if:
l n< 0
l k<0
l n £ k
Cos (Cosine)
This function takes the value of an angle in radians and returns the cosine of
the angle as a number between -1 and 1.
In a given right triangle containing an angle A, there are three sides: the side
opposite to A, the side adjacent to A, and the side opposite to the right angle
(hypotenuse). The cosine of A is the ratio of the lengths of the adjacent side
over the hypotenuse.
To convert degrees to radians, multiply by p/180. This is a single-value

function.
Syntax
Cos(Argument)
Where:
angles, in radians, for which the value of the cosine is sought. To convert
the result from radians to degrees, multiply by 180/p.

Cosh (Cosine, Hyperbolic)

Returns the hyperbolic cosine of a number. This is a single-value function.
Syntax
Cosh(Argument)
Where:
angles, in radians, for which the value of the hyperbolic cosine is sought. To
convert radians to degrees, multiply by 180/p.
Expression
Degrees (Conversion To)

Returns the value of an angle converted from radians to degrees. This is a
single-value function.
Syntax
Degrees(Argument)
Where:
Argument is a metric representing a list of values to be converted from

radians to degrees.
Example
These simple examples illustrate how the Degrees function converts an

angle entered in radians into degrees.

Function/Result Calculation
Degrees(2.27) = 130
Degrees( p/2 ) = 90
Exp (Exponent)
Returns the value e raised to the power of the number given. e is a
mathematical constant equal to 2.71828182845904. Exp is the inverse of
Ln, the natural logarithm of a number.
This is a single-value function.
For more information on the natural logarithm function, see Ln (Logarithm,

Natural), page 426.
Syntax
Exp(Argument)
Where:
power to which e is raised
Example
This simple example illustrates the results of using the Exp function. The
variable e=2.71828182845904
X = e2 ;
Exp (2) =
7.38905609893062 2.71828182845904 × 2.71828182845904 =
7.38905609893062

Factorial (Factorial)
Returns the factorial of a positive integer. The factorial of an integer N is
equal to the product of all integers from 1 to N. For example, the factorial of
5 is 1 × 2 × 3 × 4 × 5, or 120.
Syntax
Factorial(Argument)
Where:
Argument is a metric representing a list of positive integers.
Usage Notes
If a number is not an integer or is negative, the function returns a NULL.
Example
These simple examples illustrate the results of using the Factorial function.
Factorial(5) = 120 5 × 4 × 3 × 2 × 1=120
1 × 2 × 3 = 6;
Factorial(Argument) = 6,
1 × 2 × 3 × 4 × 5 × 6 × 7 = 5040;
5040, 362880 where Argument
contains the values 3,7,9 1 ×2×3×4×5×6×7×8×9
= 362880
Floor (Floor Value)

Returns the closest integer less than or equal to a given number. Floor is
often used as a rounding function. This is a single-value function.
Syntax
Floor(Argument)

Where:
Example
These simple examples illustrate how the Floor function acts on positive
and negative numbers, and how the results compare to the use of the
Ceiling function.
Argument Values Floor Ceiling
947.67 947.00 948.00
463.33 463.00 464.00
-947.67 -948.00 -947.00
-463.33 -464.00 -463.00
Int (Integer)
Returns the integer part of a number rounded down to the nearest integer.
Syntax
Int(Argument)
Where:
Usage Notes
The expression [Argument - INT(Argument)] can be used to return

the decimal part of a non-integer.
Results from Int and Trunc differ only when the values are less than 0.
Example

These simple examples illustrate how the Int function acts on positive and
negative numbers, and how the results differ from the Trunc function when
the numbers are negative.
Argument Values Int Trunc
947.67 947.00 947.00
463.33 463.00 463.00
-947.67 -948.00 -947.00
-463.33 -464.00 -463.00
Ln (Logarithm, Natural)
Returns the natural logarithm of a given number. This is a single-value
function. Ln is the inverse of the Exp function.
For more information on the Exponent function, see Exp (Exponent), page
423.
Syntax
Ln(Argument)
Where:
Argument is a metric representing a list of positive real numbers.
Example
This simple example illustrates the results of the Ln function. The variable
e=2.71828182845904.
Ln (20) = 20 = e x where x is the natural logarithm;

2.99573227355399 20 = e 2.99573227355399

Log (Logarithm)
Returns the logarithm of a given number in a specified base. This is a single-
value function.
Syntax
Log(Argument, Base)
Where:
Base is the base of the logarithm.
Example
This simple example illustrates the results of the Log function.
Log(9, 3) = 2 9 = 3 x where x is the base-3 logarithm; 9 = 3 2
Log(4, 4) = 1 4 = 4 x where x is the base-4 logarithm; 4 = 4 1
7776 = 6 x where x is the base-6 logarithm;

Log (7776, 6) = 5
7776 = 6 5
Log10 (Logarithm, Base 10)

Returns the base-10 logarithm of a given number. This is a single-value
function.
Syntax
Log10(Argument)
Where:
Example

This simple example illustrates the results of the Log10 function.
10000 = 10 x where x is the base-10

Log10(10000) = 4
logarithm; 10000 = 10 4
Mod (Modulus)
Returns the remainder of a number when divided by a another number. The
result has the same sign as the dividend. This is a single-value function.
Syntax
Mod (Argument, Divisor)
Where:
Divisor is the number used to divide the value(s) of Argument.
Usage Notes
The function returns a NULL if Divisor = 0.
Use the Quotient function to return only the integral part of the result.
Example
These simple examples illustrate the results of using the Mod function.
Mod(5, 2) = 1 5/2 = 2 remainder 1
5/-2 = 2 remainder 1 and takes the sign of

Mod(5, -2) = 1
the dividend
Mod(-5, 2) = -1 -5/2 = 2 remainder 1 and takes the sign of

the dividend
Mod(140, 2) = 0 140/2 = 70 remainder 0
Mod(83, 9) = 2 83/9 = 9 remainder 2
-5/-2 = 2 remainder 1 and takes the sign of

Mod(-5, -2) = -1
the dividend
Power
Returns the value of an input number raised to a given power. This is a
Syntax
Power(Argument, Power)
Where:
Argument is a metric representing a list of base numbers. All real numbers

are valid.
Power is the exponent to which Argument is raised.
Usage Notes
If Argument is a negative number and Power is an even number, the result is

positive; however, if Argument is negative and Power is an odd number, the
result is negative.
Example
These simple examples illustrate the results of using the Power function.

Power(5, 2) = 25 5 × 5 = 25
Power(-5, 2) = 25 -5 × -5 = 25
Power(-5, 3) = -125 -5 × -5 × -5 = -125
Power(4, 3) = 64 4 × 4 × 4 = 64
Power(20,4) =
20 × 20 × 20 × 20 = 160000
160000
Quotient
Returns the integer result of dividing two numbers. This function discards
the remainder. This is a single-value function.
Syntax
Quotient(numerator, denominator)
Where:
Numerator is a metric representing a list of real numbers to be used as the

dividend.
Denominator is a metric representing a list of real numbers to be used as

the divisor.
Usage Notes
Use the Mod function (Mod (Modulus), page 428) to retrieve only the
remainder.
Example
These simple examples illustrate the results of using the Quotient function.

Quotient(5, 2) = 2 5/2 = 2 remainder 1
Quotient(83, 9) = 9 83/9 = 9 remainder 2
Radians (Conversion To)

Returns the value of an angle converted from degrees to radians. This is a
Syntax
Radians(Argument)
Where:
Argument is a metric representing a list of values to convert from degrees

to radians.
Example
This simple example illustrates how the Radians function converts an angle
entered in degrees into radians.
Radians(130) = 2.27
Randbetween (Random Number Between Two Values)

Returns a random number that falls between two specified values. The value
returned is a real number, as opposed to an integer value.
A different number is returned each time the function is run. However, if the
Randbetween function is used to modify multiple rows of values on a report,
be aware that if you use a simple definition such as Randbetween(2, 10) the
same random number will be applied to each row of data on the report. You

can apply a different random number to each row of data in a report by

including metric values in the Randbetween calculation, and an example of
this is provided below.
Syntax
Randbetween(Bottom, Top)
Where:
Bottom is a real number or metric representing a list of real numbers that

are the lowest value(s) the function can return; the bottom of the range. You
can also include expressions that result in a real number value.
Top is a real number or metric representing a list of real numbers which are
the highest value(s) the function can return; the top of the range. You can
also include expressions that result in a real number value.
While Randbetween can accept real values as input values for Bottom and
Top, the input values are processed differently based on the following:
l Processed by the Analytical Engine: Functions provided with

MicroStrategy can be evaluated by the MicroStrategy Analytical Engine in
cases where functions cannot be evaluated by a database, or when
features such as derived metrics use a function as part of their definition.
When Randbetween is processed by the Analytical Engine, both input
values are rounded down to the nearest integer value before determining a
random number. For example, the expression Randbetween(0.5,1.6)
for a derived metric will first round both input values down, making the
resulting expression Randbetween(0,1). A random, real number is then
returned between the values of 0 and 1.
You can determine if a metric is processed by the Analytical Engine by

looking at the SQL view of a report.
l Processed by a database: Functions provided with MicroStrategy can be

passed to the database for processing when the database supports that

function. See Appendix A, MicroStrategy and Database Support for

Functions for a list of the databases and the functions they support.
When Randbetween is processed by a database, the real values can be

used directly as input values without any rounding. For example, the
expression Randbetween(0.5,1.6) for a metric will return a random,
real number between the values of 0.5 and 1.6.
Example
Random numbers are used often in statistical analysis. A common practice

is to apply a random percentage to another value. To do this, you can use
the Randbetween function to return a number between 0 and 1. This can be
achieved with the following syntax:
Randbetween(0, 1)
This returns a real number value between zero and one. This can be used in
an expression to multiply against another value, which would return a
random percentage of that value. For example, you can return a random
percentage of revenue by creating a metric with the following definition:
(Sum(Revenue) {~+} * Randbetween(0, 1))
When using a metric with this definition on a report the same random
percentage is applied to all revenue values on a report.
In the definition provided below, the metric Revenue is used in the

Randbetween calculation. By including the revenue data as part of the
calculation of Randbetween, a different random number is generated for
each row of revenue data on a report.
Randbetween(0, Revenue)
These two metrics are created as Random Revenue 1 and Random Revenue
2 respectively, and displayed on the same report shown below:

Notice that the metric Random Revenue 2, which includes the revenue data
as part of the Randbetween calculation, is using a different percentage for
each row of revenue data. However, the metric Random Revenue 1 which
uses a simple definition of the Randbetween calculation as Randbetween
(0,1) uses the same random percentage to modify all rows of Revenue
data.
Round (Round to Nearest Integer)

Returns the input values rounded to the nearest integer. This is a single-
value function.
Syntax
Round(Argument)
Where:
Example
These simple examples illustrate the result of using the Round function, and
compare them to the Int function results for the same values.
Argument Values Round Int
947.67 948.00 947.00
463.33 463.00 463.00
-947.67 -948.00 -948.00
-463.33 -463.00 -464.00

Round2 (Round to Specified Precision)

Returns the input values rounded to a specified number of digits. This is a
Syntax
Round2(Argument, Precision)
Where:
Argument is a metric representing a list of real numbers
Precision is an integer value specifying the placement of the digits to

which the Argument value(s) are rounded.
l If Precision = 0, then the values are rounded to the nearest integer. This
has the same results as using the Round function.
l If Precision is greater than 0, then the values are rounded to the

decimal place indicated.
l If Precision is less than 0, then the values are rounded to the specified
digits to the left of the decimal and starting to the left of the integer in the
ones place.
Example
This simple example illustrates how the Precision value affects the results of
the Round2 function.
Round2 Round2 Round2

Argument values
(argument, 0) (argument, 1) (argument, -1)
947.67 948.00 947.70 950.00
811.00 811.00 811.00 810.00
463.33 463.00 463.30 460.00

Sin (Sine)
This function takes the value of an angle in radians and returns the sine of
the angle as a number between -1 and 1.
(hypotenuse). The sine of A is the ratio of the lengths of the opposite side
over the hypotenuse.

function.
Syntax
Sin(Argument)
Where:
Argument is a metric representing a list of values. The values are angles, in

radians, for which the sine is sought. Multiplying the argument by p/180
converts an angle from degrees to radians.
Sinh (Sine, Hyperbolic)

Returns the hyperbolic sine of a given angle. The value of the hyperbolic
sine is used to approximate a cumulative probability distribution. This is a
Syntax
Sinh(Argument)
Where:
Argument is a metric representing a list of values. The values are angles, in

radians, for which the hyperbolic sine is sought. Multiplying the argument by
p/180 converts an angle from degrees to radians.
Expression

Sqrt (Square Root)

Returns the square root of a given positive number. This is a single-value
function.
Syntax
Sqrt(Argument)
Where:
Usage Notes
A NULL is returned if an argument value is less than 0.
Example
This simple example illustrates the results of using the Sqrt function.
Sqrt(25) = 5
Sqrt(169) = 13
Tan (Tangent)
This function takes the value of an angle in radians and returns the tangent
of the angle.
(hypotenuse). The tangent of A is the ratio of the lengths of the opposite side
over the adjacent side.

function.
Syntax

Tan(Argument)
Where:
Tanh (Tangent, Hyperbolic)

Returns the value of the hyperbolic tangent of a given number. This is a
Syntax
Tanh(Argument)
Where:
Expression
Trunc (Truncate)
Returns the integer component of a given number. The function removes the
fractional part of a real number so that only the integer part remains. The
fractional part is removed without rounding the number up or down. This is a
Syntax
Trunc(Argument)
Where:
Usage Notes
Results from Trunc and Int differ only when the values are less than 0.
Example

This simple example illustrates how the Trunc function acts on positive and
negative numbers, and how it differs from the Int function.
Argument Values Int Trunc
947.67 947.00 947.00
463.33 463.00 463.00
-947.67 -948.00 -947.00
-463.33 -464.00 -463.00
Statistical Functions
The statistical functions include a wide range of functions designed to
provide you with the tools to perform statistical analysis on your data.
AvgDev (Average Deviation)

Returns the average of the absolute deviations of a set of data points from
the mean. Similar to standard deviation, average deviation is a measure of
the variability in a data set.
Syntax
AvgDev <FactID> (Argument)
Where:

Argument is a fact or metric representing a list of numbers.
Expression

In this expression, x is a value in a fact table.
Where:
is the average of all x i
BetaDistribution
Returns the cumulative Beta distribution of the variables alpha and beta,
where the probability density function is given by
Where:
This function is used to study percentage variations for an item across

population samples (for example, how much of a day people spend watching
television).
Syntax
BetaDistribution < Lower Bound, Upper Bound > (x, alpha, beta)
Where:
x is the value between Lower Bound (a) and Upper Bound (b) at which
the function is evaluated.

alpha and beta are distribution parameters.
l alpha =
l beta =
Lower Bound is an optional lower bound of the interval of x.
Upper Bound is an optional upper bound of the interval of x.
Usage Notes
The following conditions are invalid:
l x, alpha, or beta is nonnumeric.
l
alpha £ 0 or beta £ 0.
l x < Lower Bound or x > Upper Bound.
l Lower Bound = Upper Bound.
BinomialDistribution
Returns either the binomial probability or the cumulative binomial probability
of variables n and p. This function is used in cases in which:
l There is a fixed number of tests
l Each test can only result in either success or failure (binary)
l Tests are independent of all other tests
l The probability of success is constant, for example, to calculate the

probability that two out of the next three babies will be male
Syntax
BinomialDistribution <Cumulative> (x, n, p)
Where:
x is the number of successes.
n is the number of tests.

p is the number indicating the probability of a success in each trial.
Cumulative is a logical value that determines the form of the function.
l If Cumulative=0, it computes the Cumulative Binomial probability.
l If Cumulative=1, it computes the Binomial probability.
Expression
For the binomial probability:
Where:
is Combine(n, x).
The cumulative binomial distribution is:
Usage Notes
If x or n is not an integer, it is truncated.
l One of the arguments is nonnumeric.
l x>n
l x< 0
l p < 0 or p > 1

ChiSquareDistribution
Returns the right-tailed chi-squared distribution of the degrees of freedom,
where the probability density function is given by
Where:
is the degree of freedom
is the gamma function
The chi-squared distribution is associated with a chi-squared test. A chi-

squared test is used to compare observed values against expected ones. It
determines the validity of a hypothesis, such as the prediction that the next
generation of a certain plant hybrid will exhibit a specific combination of
colors.
Syntax
ChiSquareDistribution(x, df)
Where:
x is a non-negative decimal value at which the distribution is evaluated.
df is an integer greater than or equal to one (>=1) indicating the number of

degrees of freedom for the distribution.
Usage Notes
If df is not an integer, it is truncated

l x < 0
l df < 1
ChiSquareTest (Chi-Square Test for Goodness of Fit)

Returns the result of a test for goodness of fit. This function returns the chi-
squared distribution for a statistical value and its associated degrees of
freedom to determine, for example, to what extent an experiment verifies
hypothesized results.
Reality: H 0 is Reality: H 1 is
Decision
correct correct
Accept H 0 No problem Type 1 error
Reject H 0 Type 2 error No problem
Definition
P-value is a probability of making a Type 2 error.
ChiSquareTest returns the p-value for the hypothesis test in the following
form:
l H 0 : Argument1 and Argument2 come from the same distribution.
l H 1 : Argument1 and Argument2 come from different distributions.
Syntax
ChiSquareTest(Argument1, Argument2)
Where:

Argument1 is a fact or metric representing a list of values that contains

observed values, which are to be tested against expected values.
Argument2 is a fact or metric representing a list of expected values, which

are compared against the observed values provided by Argument1.
Usage Notes
This function returns the probability for a chi-squared (x 2 ) statistic and its
associated degrees of freedom (df).
This function results in an invalid value if Argument1 and Argument2 have

a different number of data points. Meaning, each observed value must have
an expected value to be compared to.
Confidence (Confidence Interval)

Returns the confidence interval for a population mean. The confidence
interval is a range of values on either side of a mean. For example, if you
order a product through the mail, you can determine, within a given level of
confidence, the earliest and latest date of expected arrival.
Syntax
Confidence(Alpha, Stdev, Size)
Where:
Alpha is the level of significance at which confidence is computed.
Stdev is the standard deviation of the population for the given data range.
Size is a positive integer indicating the size of the sample.
Expression
The confidence level is expressed as [100(1-alpha)%]. If alpha = 0.05, then:
l Confidence = 95%
l The area equivalent to 95% of the population = 1.96

l The confidence interval is
Usage Notes
If size is not an integer, it is truncated.
l alpha < 0 or alpha > 1
l stdev < 0
l size < 1
Example
For an example of the Confidence function, see Confidence Level Example,

page 91.
Correlation
A correlation of two values describes the degree to which the values are
related or associated. Values that are closely related with either a positive
or negative correlation will have a correlation close to 1 or -1 respectively,
while values that are not correlated at all will have a correlation close to 0.
To determine the correlation of two values while applying a weight to the

comparisons, see WeightedCorr (Weighted Correlation), page 254.
This function is the same as Pearson (Pearson Product Moment Correlation

Coefficient), page 479.
Syntax
Correlation <FactID> (Argument1, Argument2)
Where:


Argument1 and Argument2 are lists of values represented by facts or

metrics.
Usage Notes
The correlation coefficient is measured on a scale that varies from + 1

through 0 to - 1. So only a value between -1 and 1 is returned.
Complete correlation between two variables is expressed by either + 1 or -1.

When one variable increases as the other increases, the correlation is
positive; when one decreases as the other increases, the correlation is
negative. Complete absence of correlation is represented by 0.
If an array or reference argument contains text, logical values, or empty

included.
If Argument1 and Argument2 have a different number of data points,

Correlation returns an error.
If either Argument1 or Argument2 is empty, or if the standard deviation of

their values equals zero, Correlation returns an error.
Covariance
Covariance is used to examine the relationship between two data sets. For
instance, the covariance could be used to examine whether an increase in
income is related to higher education levels. A covariance greater or less
than zero indicate a relationship, while a value of zero indicates no
relationship.
To determine the covariance of two values while applying a weight to the

comparisons, see WeightedCov (Weighted Covariance), page 257.
Covariance is similar to Correlation and Pearson, which also describe

the relationship between two data sets. While Covariance describes the

relationships with positive or negative values, Correlation and Pearson

normalize these positive or negative values to a range between -1 and 1.
Syntax
Covariance <FactID> (Argument1, Argument2)
Where:
Argument1 and Argument2 are facts or metrics representing lists of

values.

Expression
In the expression that follows:
l Argument1 = X
l Argument2= Y
Where:
= average value of x
= average value of y
Usage Notes
l Argument1 and Argument2 do not contain the same number of data points.
l Either value set is empty.

CritBinomial (Criterion Binomial)

Returns the smallest value for which the cumulative binomial distribution is
greater than or equal to a criterion value. Use this function for quality
control, for example, to determine the maximum number of defective items
that can come off an assembly line run before rejecting an entire lot.
Syntax
CritBinomial(Trials, Probability_s, Alpha)
Where:
Trials is the number of Bernoulli trials.
Probability_s is the probability of a success on each trial.
Alpha is the criterion value.
Usage Notes
l Any argument is nonnumeric.
l Trials is not an integer.
l Trials < 0
l Probability_s is < 0 or probability_s > 1
l alpha < 0 or alpha > 1
ExponentialDistribution
Returns either the probability density or the cumulative distribution of a
random variable. This function is used to model the time between events.
For example, it can be used to estimate the probability that an automatic-
teller machine will take no more than one minute to deliver requested cash.
Syntax
ExponentialDistribution <Cumulative> (x, lambda)

Where:
Cumulative is a logical value that determines the form of the function:
l If Cumulative=0 (default), it returns the cumulative distribution.
l If Cumulative=1, it returns the probability density.
x is the value of the function.
lambda is the value of the parameter.
Expression
For probability density:
For cumulative distribution:
Usage Notes
l Either x or lambda is nonnumeric
l x<0
l
lambda £ 0
Fisher (Fisher Transformation)

Returns the Fisher transformation at x. The Fisher transformation yields a
normally-distributed function, as opposed to a skewed one. It is used to
perform hypothesis-testing on the correlation coefficient.
Syntax
Fisher (x)

Where:
x is the number for which the transformation is sought.
Expression
Usage Notes
l x is nonnumeric.
l
x £ -1 or x ³ 1.
FDistribution (F-Probability Distribution)

Returns the right-tailed distribution (F-probability) of a random variable. This
function is used to determine the difference, in degrees of diversity, between
two data sets. For example, test scores administered to males and females
entering high school can be examined using this function to determine
whether the variability in females is different from that found in males.
Syntax
FDistribution(x, df1, df2)
Where:
x is the value at which the function is evaluated.
df 1 represents the degrees of freedom of numerator values.
df 2 represents the degrees of freedom of denominator values.
Expression
Where:

F is a random variable that has an F distribution.
Usage Notes
If df 1 or df 2 is not an integer, it is truncated.
l An argument is non-numeric
l x<0
l df 1 < 1
l df 2 < 1
Forecast
The Forecast function uses data recorded for certain time periods to
calculate and display the expected value of data for an unknown time period.
This is most commonly used to predict future metric values based on
historical data. For example, based on revenue data for three years, you
want to predict what your Revenue will be for the coming year.
This function uses a linear equation to predict future values. Therefore, it is

best suited to predict values that increase or decrease in a linear trend. For
example, if your profits are increasing by $10,000 each year, this is a
scenario that is well suited for a prediction based on a linear equation.
For a function that predicts values in a similar way for data that is increasing
or decreasing exponentially, see Growth, page 458.
Syntax
Forecast <FactID>(FutureTime, KnownData1,...,KnownDataN,

KnownTimeIntervals1,...,KnownTimeIntervalN)
Where:
FutureTime is the point in time for which to predict the data. This is
commonly defined as an integer. For example, to predict values for the year

2009, you would enter the value 2009 for FutureTime (this assumes your
year data is defined with IDs that match the year).
KnownData1,...,KnownDataN are separate scalar values that provide

the data to use to predict a future value for. While you can use Forecast to
enter these values manually, it is more common to use ForecastV which
can use a metric to enter a list of values, as described in ForecastV
(Forecast, Vector Input), page 453.
KnownTimeIntervals1,...,KnownTimeIntervalN are separate scalar

values that provide the time intervals for which data is known. While you can
use Forecast to enter these values manually, it is more common to use
ForecastV which can use an attribute to enter a list of time intervals, as
described in ForecastV (Forecast, Vector Input), page 453.

Example
As described in the syntax requirements, you can use Forecast to enter

values and time intervals manually. However, it is more common to use
ForecastV to use metrics and attributes to enter this information, as
described in ForecastV (Forecast, Vector Input), page 453.
ForecastV (Forecast, Vector Input)

The ForecastV function uses data recorded for certain time periods to
historical data. For example, based on revenue data for three years, you can
predict what your Revenue will be for the coming year.
You can use ForecastV to predict values for any point in time, including
dates that have known data. However, this type of function is meant for
predicting future, unknown values.


best suited to predict values that are increasing or decreasing in a linear
trend. For example, if your profits are increasing by $10,000 each year, this
is a scenario that is well suited for a prediction based on a linear equation.
or decreasing exponentially, see GrowthV (Growth, Vector Input), page 459.
Syntax
ForecastV <FactID>(FutureTime, KnownData, KnownTimeIntervals)
Where:
FutureTime is the point in time for which to predict the data. This is
commonly defined as an integer. For example, to predict values for the year
2009, you would enter the value 2009 for FutureTime (this assumes your
year data is defined with IDs that match the year).
KnownData is an object representing a list of values. This list of values is

the known data to use to predict a future value for. KnownData is commonly
used for metrics that represent data such as revenue, profit, and so on.
KnownTimeIntervals is an object representing a list of values. This list

of values provide the time intervals for which data is known.
KnownTimeIntervals is commonly used for attributes that represent
intervals of time such as year, quarter, month, and so on.
Since this function allows you to enter a list of time intervals, it can account
for missing time intervals more accurately than the TrendV function (see
TrendV (Trend, Vector Input), page 488). For an example, see the examples
described below.

Example

The ForecastV function is useful to display predicted future values of a

metric as a subtotal of the metric on a report.
For example, in the MicroStrategy Tutorial project a new subtotal named

2008 Forecast can be created with the following definition:
ForecastV(2008, x, Year)
Since this is in a subtotal, the x is a placeholder for a metric. When a

subtotal is applied to a metric on a report, the placeholder inputs that
metric’s values.
Also, 2008 is used for the period in time to predict metric data for because
the Year attribute uses ID values that match the year.
After making the new custom subtotal available for the Revenue metric (see
the Advanced Reporting Guide for steps to make custom subtotals available
for metrics) a report with the attribute Year and the metric Revenue is
created. When the 2008 Forecast is displayed on the report, it displays the
predicted value for 2008 Revenue, as shown below:
Since the ForecastV function allows you to enter a list of time intervals, it
can account for time intervals that you have no data for. For example,
consider the same report in a scenario in which there was no data available
for the year 2006. The ForecastV function can still use the two available
years to predict the revenue for 2008.

FTest
Returns the P-value of an F-test on the variances of two sets of data. It tests
that the variances in the two sets are not significantly different. For example,
given the test scores from public and private schools, this function enables
you to determine whether diversity levels between the two school types are
significantly different.
Decision Reality: H 0 is Reality: H 1 is

correct correct
Definition
P-value is basically a probability of making a Type 2 error.
FTest returns the p-value for the hypothesis test in the following form:
l :
l :

Syntax
FTest <Hypothesis type, Ratio> (Argument1, Argument2)
Usage Notes
l Argument1 and Argument2 contain a different number of data points
l The variance of either data set is zero
GammaDistribution
Returns either the cumulative Gamma distribution or the probability density
of a random variable. This function is used in queueing analysis to study
variables that may have a skewed distribution.
Syntax
GammaDistribution <Cumulative> (x, Alpha, Beta)
Where:
x is the value at which you want to evaluate the distribution.
Alpha is a distribution parameter.
Beta is a distribution parameter.
l If Cumulative=1, it returns the probability density.
Expression

The expression for the standard gamma distribution is:
If alpha = 1, this function returns the exponential distribution as:
Usage Notes
When alpha > 0, this function is also known as the Erlang distribution.
l Either alpha or beta is nonnumeric
l
Either alpha £ 0 or beta £ 0
l x<0
Growth
The Growth function uses data recorded for certain time intervals to
historical data. For example, based on revenue data for three years, you can
This function uses an exponential equation to predict future values.

Therefore, it is best suited to predict values that increase or decrease in an
exponential trend. For example, if your profits are increasing by 4 percent
each year, this is a scenario that is well suited for a prediction based on an
exponential equation.

or decreasing with a linear trend, see Forecast, page 452 or Trend, page
487.
Syntax
Growth <FactID>
(KnownData1,...,KnownDataN, FutureTimeOffset)
Where:
KnownData1,...,KnownDataN are separate scalar values that provide the

data to use to predict a future value for. While you can use Growth to enter
these values manually, it is more common to use GrowthV which can use a
metric to enter a list of values, as described in GrowthV (Growth, Vector
Input), page 459.
FutureTimeOffset is the point in time after the final time interval

available for which to predict the data. This is commonly defined as an
integer. For example, to predict values for the year 2009 when the last
available data you have is for 2007, you would enter the value of 2 for
FutureTimeOffset.

Example
As described in the syntax requirements, you can use Growth to enter

values manually. However, it is more common to use GrowthV to use
metrics and attributes to enter this information, as described in GrowthV
(Growth, Vector Input), page 459.
GrowthV (Growth, Vector Input)

The GrowthV function uses data recorded for certain time intervals to

You can use GrowthV to predict values for any point in time, including
dates that have known data. However, this type of function is meant for
predicting future, unknown values.
This function uses an exponential equation to predict future values.

Therefore, it is best suited to predict values that increase or decrease in an
exponential trend. For example, if your profits are increasing by 15 percent
each year, this is a scenario that is well suited for a prediction based on a
linear equation.
or decreasing with a linear trend, see ForecastV (Forecast, Vector Input),
page 453 or TrendV (Trend, Vector Input), page 488.
Syntax
GrowthV <FactID, SortBy>(KnownData, FutureTimeOffset)
Where:


FutureTimeOffset.
The GrowthV function assumes that data is available for sequential time
intervals with no missing data. If your data meets these requirements, this
allows you to use a single expression to use on reports with varying time
intervals such as year, month, quarter, and so on. For an example, see the
examples described below.



information, see BreakBy and SortBy Parameters, page 27.
Example
The GrowthV function is useful to display predicted future values of a metric

as a subtotal of the metric on a report.

2008 Growth can be created with the following definition:
GrowthV(x, 1)

metric’s values.
Also, 1 is used as the offset in the time interval to predict data for the first
time interval that is not known.
created. When the 2008 Growth is displayed on the report, it displays the
This predicted value is different than the values predicted by the functions
ForecastV and TrendV, as described in ForecastV (Forecast, Vector
Input), page 453 and TrendV (Trend, Vector Input), page 488. This
difference is due to GrowthV predicting values based on an exponential
trend, and as such should be used with data that reflects such a trend.

HeteroscedasticTTest and HomoscedasticTTest

Return the P-value for the hypothesis test. These two functions are used to
determine the level of variance between the means of paired samples,
assuming both samples have different arguments. For example, they may be
used when a given group is to be tested before and after an experiment.

correct correct
Definition
P-value is a probability of making a Type 2 error.
HeteroscedasticTTest or HomoscedasticTTest returns the p-value for the

hypothesis test in the following form:
l For H 0 :
l For H 1 :

Syntax
HeteroscedasticTTest or HomoscedasticTTest <Hypothesis type,

offset> (Argument1, Argument2)
Usage Notes
Heteroscedastic t-tests are based on the assumption that variances between

two sample data ranges are unequal [σ 2 (Argument1) ¹ σ 2 (Argument2)].
Homoscedastic t-tests are based on the assumption that variances between

two sample data ranges are equal [σ 2 (Argument1) = σ 2 (Argument2)].
l Argument1 and Argument2 have a different number of data points, and

Hypothesis type = 1 (paired).
l Offset or Hypothesis type is nonnumeric.
Example
For an example using both Heteroscedastic T-test and Homoscedastic T-

test, see Hypothesis Testing Example, page 83.
HypergeometricDistribution
Returns the probability of a given number of sample successes, given the
sample size, population successes, and population size. Use this function
for problems with a finite population, where each observation is either a
success or a failure, and where each subset of a given size is chosen with
equal likelihood.
Syntax
HypergeometricDistribution(x, n, M, N)

Where:
x is the number of successes in the sample.
n is the size of the sample.
M is the number of successes in the population.
N is the population size.
Expression
Usage Notes
All arguments are truncated to integers.
l Any argument is nonnumeric
l x < 0 or x is greater than the lesser of n or M
l x is less than the larger of 0 or (n - N + M)
l n < 0 or n > N
l M < 0 or M > N
l N<0
Intercept
Returns the point at which a line intersects the y-axis in a graph. This
function is used to determine the value of a dependent variable Y when the
value of its associated independent variable X is 0. The point of intersection
is based on a best-fit regression line plotted through available values of X
and Y.

Syntax
Intercept <FactID> (Argument1, Argument2)
Where:
Argument1 is the dependent set of observation data represented by a fact

or metric.
Argument2 is the independent set of observation data represented by a fact

or metric.

Expression
In the expressions that follow, Argument1 = Y, Argument2 = X.
The equation for the intercept of the regression line is
The slope of the regression line is calculated as:
Usage Notes
This function results in an invalid value if Argument1 and Argument2 contain

a different number of data points.
Example
For an example using the Intercept function, see Hypothesis Testing

Example, page 83.

InverseBetaDistribution (Inverse of the Beta Distribution)

Returns the inverse of the cumulative Beta distribution.
For information on the Beta distribution function, see BetaDistribution, page

440.
Syntax
InverseBetaDistribution <Lower Bound, Upper Bound> (x, Alpha,

Beta)
Where:
x is a probability associated with the beta distribution.
Alpha and Beta are distribution parameters.
Lower Bound is an optional lower bound of the interval of x.
Upper Bound is an optional upper bound of the interval of x.
Usage Notes
l x, alpha, or beta is nonnumeric
l
alpha £ 0 or beta £ 0
l
x £ 0 or x ³ 1
l x < Lower Bound or x > Upper Bound
l Lower Bound > Upper Bound
InverseChiDistribution (Inverse of Chi-Squared Distribution)

Returns the inverse of the given right-tailed Chi-squared distribution greater
than or equal to a pre-specified value. This function is used to compare
observed results against expected ones to assess the validity of a
hypothesis.

For information on the Chi-squared function, see ChiSquareDistribution,

page 443.
Syntax
InverseChiDistribution(x, df)
Where:
x is the probability associated with a given chi-squared distribution.
df is the number of degrees of freedom for the distribution.
Usage Notes
If df is not an integer, it is truncated
l x < 0 or x >1
l df < 1
InverseFisher (Inverse of the Fisher Transformation)

Returns the inverse of the Fisher transformation. This function is used to
analyze correlations between data arrays.
For information on the Fisher transformation function, see Fisher (Fisher

Transformation), page 450.
Syntax
InverseFisher(x)
Where:
x is the value for which the inverse of the Fisher transformation is

performed.
Expression

Usage Notes
This function results in an invalid value if x is nonnumeric.
InverseFDistribution (Inverse of F-Probability Distribution)

Returns the inverse of the F probability distribution. This function is used to
return critical values from the F distribution. For example, the output of an
analysis-of-variation (anova) calculation may include data for F-statistic, F-
probability, and F-critical values at the 0.05 level of significance. In such
cases, the significance level can be used as the probability argument for
FINV to return the critical value of F.
For information on the F probability distribution function, see FDistribution

(F-Probability Distribution), page 451.
Syntax
InverseFDistribution (x, df1, df2)
Where:
x is the probability associated with the F right-tailed distribution.
df 1 represents the degrees of freedom of numerator values.
df 2 represents the degrees of freedom of denominator values.
Usage Notes
If df 1 or df 2 is not an integer, it is truncated.
l An argument is nonnumeric
l x< 0 or x > 1

l df 1 < 1
l df 2 < 1
InverseGammaDistribution (Inverse of Gamma Distribution)

Returns the inverse of the cumulative Gamma distribution.
For more information on the cumulative Gamma distribution function, see

GammaDistribution, page 457.
Syntax
InverseGammaDistribution (x, Alpha, Beta)
Where:
x is the value at which you want to evaluate the distribution.
Usage Notes
l Either alpha or beta is nonnumeric
l
Either alpha £ 0 or beta £ 0
l x < 0 or x > 1
InverseLognormalDistribution (Inverse of Lognormal

Distribution)
Returns the inverse of the lognormal cumulative distribution. Ln(x) is
normally distributed using mean and standard deviation as parameters.
For more information on the lognormal cumulative distribution function, see

LognormalDistribution, page 473.
Syntax
InverseLognormalDistribution(x, Mean, Stdev)

Where:
x is the probability associated with the lognormal distribution.
Mean is the mean of ln(x).
Stdev is the standard deviation of ln(x).
Expression
In the expression that follows, mean = m, stdev =s.
Usage Notes
l One of the arguments is nonnumeric
l x<0
l x>1
l stdev < 0
InverseNormDistribution (Inverse of Normal Cumulative

Distribution)
Returns the inverse of the normal cumulative distribution for a given mean
and standard deviation.
For more information on the normal cumulative distribution function, see

NormalDistribution (Normal Cumulative Distribution), page 476.
Syntax
InverseNormDistribution(x, Mean, Stdev)
Where:

x is the probability corresponding to a given normal distribution.
Mean is the arithmetic mean of the distribution.
Stdev is the standard deviation of the distribution.
Usage Notes
If mean = 0 and stdev = 1, this function returns the standard normal

distribution
l stdev £ 0.
InverseNormSDistribution (Inverse of Standard Normal

Cumulative Distribution)
Returns the inverse, or critical value, of the cumulative standard normal
distribution. This function computes the critical value so that the cumulative
distribution is greater than or equal to a pre-specified value.
For more information on the cumulative standard normal distribution

function, see StandardNormalDistribution (Standard Normal Cumulative
Distribution), page 485.
Syntax
InverseNormSDistribution(x)
Where:
x is the probability corresponding to the normal distribution.
Usage Notes
l x is nonnumeric.
l x < 0 or x > 1.

InverseTDistribution (Inverse of T-Distribution)

Returns the inverse of the right-tailed T-distribution.
For more information on the right-tailed T-distribution function, see

TDistribution, page 486.
Syntax
InverseTDistribution(Probability, df)
Where:
Probability is the probability associated with the t-distribution.
df is a value that indicates the degrees of freedom of the distribution.
Expression
Where:
X is a random value that follows the t-distribution.
Usage Notes
If df is not an integer, it is truncated.
l Either probability or df is nonnumeric
l Probability < 0 or probability > 1
l df < 1
Kurtosis
Returns the kurtosis, which is a measure of whether the data is heavy or
light-tailed relative to a normal distribution. Datasets with high kurtosis tend
to have heavy tails, or outliers. Datasets with low kurtosis tend to have light
tails, or lack of outliers.

Syntax
Kurtosis <FactID> (Argument)
Where:
Argument is a set of values represented by a fact or metric.

Expression
In the expression that follows, s is the sample standard deviation.
LognormalDistribution
Returns the lognormal cumulative distribution of a random variable X, where
the logarithm of X is normally distributed. LognormalDistribution takes
mean and standard deviation as parameters. This function is used to
analyze data that has been logarithmically transformed.
Syntax
LognormalDistribution(x, Mean, Stdev)
Where:
x is the value at which the function is evaluated.
Mean is the mean of ln(x).
Stdev is the standard deviation of ln(x).
Expression
In the expression that follows, mean = m, stdev = s.
Usage Notes

l One of the arguments is nonnumeric
l
x £0
l
stdev £ 0
MeanTTest (Mean T-test)

Calculates the P-value of a T-test that tests the mean of the data against a
particular value. This function is used to determine the closeness in the
value of the means of two samples coming from the same population.

correct correct
Definition
MeanTTest returns the p-value for the hypothesis test in the following form:
l For H 0 :
l For H 1 :

Syntax
MeanTTest <Hypothesis type> (Argument, Hypothesis mean)
Where:
Argument is the value or list of values, represented by a fact or metric, that

contains sample data.
Hypothesis mean is m 0 .
Hypothesistype denotes the type of t-test to be performed. Hypothesis

type can be -1, 0, 1, or 2.
NegativeBinomialDistribution
This function determines the probability that there will be F failures before
the Nth success (S) with a constant probability of success P. The negative
binomial distribution is similar to the binomial distribution, but the number of
successes is fixed and the number of trials is variable.
As in the binomial distribution, trials are assumed to be independent. For

example, to find ten individuals with excellent reflexes, knowing that the
probability that a candidate has these qualifications is 0.3, this function
calculates the probability that a certain number of unqualified candidates
will be interviewed before the ten people sought are selected.
Syntax
NegativeBinomialDistribution(F, S, P)
Where:
F is the number of failures.
S is the threshold number of successes.
P is the probability of a success.

Expression
In the expression that follows, F = x, S = r.
Usage Notes
If either F or S is not an integer, it is truncated.
l P< 0 or P> 1.
l
F + (S - 1) £ 0.
NormalDistribution (Normal Cumulative Distribution)

Calculates the normal distribution of the mean and standard deviation of a
set of values. Returns either the cumulative distribution or the probability
density. This function is widely applied in statistics, including in the area of
hypothesis testing.
Syntax
NormalDistribution <Cumulative> (x, Mean, Stdev)
Where:
x is the value for which the distribution is sought.
StDev is the standard deviation of the distribution.
l If Cumulative=1 it returns the probability density.

Expression
In the expression that follows, z = x, mean = m, stdev = s.
Usage Notes
l Either mean or stdev is nonnumeric
l Stdev £ 0
If mean = 0 and stdev = 1, the function returns the standard normal

distribution.
PairedTTest (Paired T-test, Two-Sample for Means)

Calculates the P-value of a T-test on the means of two paired sets of data.
Two-sample analyses are used to determine the level of variance between
the means of paired samples.

correct correct
Definition

PairedTTest returns the P-value for the hypothesis test in the following
form:
l H0:
l H1:
Syntax
PairedTTest <Type, Offset> (Argument1, Argument2)
Where:
Argument1 is a fact or metric representing a set of sample values.
Argument2 is a fact or metric representing a set of sample values.
Type is a logical value that determines the form of the function.
Usage Notes
One of the results of this sample is pooled variance, an accumulated

measure of spread data about the mean. Pooled variance is calculated as:

Pearson (Pearson Product Moment Correlation Coefficient)

This function returns the Pearson product moment correlation coefficient of
a set of values. The coefficient is a dimensionless index that ranges from -
1.0 to 1.0 (inclusive). It is used to determine the extent of a relationship
between two data sets.
Syntax
Pearson <FactID> (Argument1, Argument2)
Where:
Argument1 is a list of independent values represented by a fact or metric.
Argument2 is a list of dependent values represented by a fact or metric.

Expression
In the expression that follows, Argument1 = X, Argument2 = Y.
Usage Notes
l Either Argument1 or Argument2 contains one or more null values
l Argument1 and Argument2 contain a different number of data points
Permut (Permutation)
Returns the number of possible permutations for a given number of items
selected as a group from a set.In permutation calculations, the internal order
of the items selected affects the return. This differs from calculations of

combinations, in which the order of the items selected does not affect the
return. Combination (Combine) is categorized as a mathematical function.
Syntax
Permut(Number, NumberChosen)
Where:
Number is the number of items in a set.
NumberChosen is the number of items selected for permutation.
Expression
In the expression that follows, Number = n; NumberChosen = k.
Usage Notes
This function returns an error value if:
l Number < 0
l NumberChosen < 0
l Number < NumberChosen
PoissonDistribution
Returns the Poisson distribution of a random variable. This function predicts
the number of identified events to take place over a specified period. For
example, it can predict the number of cars to arrive at a toll plaza in a one-
minute interval.
Syntax

PoissonDistribution <Cumulative> (x, lambda)
Where:
x is the number of events in the prediction.
lambda is the expected numeric value of the expression.
Cumulative is a logical value that determines the form of the probability

distribution returned:
l If Cumulative = 0 (default), the function returns the probability that the

number of random events will be between 0 and x.
l If Cumulative = 1, the function returns the probability that the number of

random events will be exactly x.
Expression
For type = 1:
For type = 0:
Usage Notes
If x is not an integer, it is truncated.
l x or lambda is nonnumeric
l x < 0
l lambda £ 0

RSquare (Square of Pearson Product Moment Correlation

Coefficient)
Returns the square of the Pearson product moment correlation coefficient
for a set of values. RSquare gives the proportion of the variance in y that is
attributable to the variance in x.
Syntax
RSquare <FactID> (Argument1, Argument2)
Where:
Argument1 is a list of data-point coordinates.
Argument2 is a list of data point coordinates.

Usage Notes
l Argument1 and Argument2 have a different number of values.
l Either Argument1 or Argument2 contains one or more NULL values.
Skew
Returns the skewness of a data set in a given list of values. Skewness is the
degree of asymmetry of a distribution in relation to its mean. A positive
skewness indicates asymmetry biased towards positive values, while a
negative skewness indicates the opposite.
Syntax
Skew <FactID> (Argument)
Where:
Argument is a fact or metric representing a list of values.


Expression
In the expression that follows, s is the sample standard deviation.
Slope (of a Linear Regression)

Returns the slope of a linear regression line. In a regression line passing
through a set of data points in data sets Argument1 and Argument2, the
slope is the vertical distance divided by the horizontal distance between any
two points on the line. This ratio is also known as the rate of change along
the line.
Syntax
Slope <FactID> (Argument1, Argument2)
Where:
Argument1 is a fact or metric containing numerically dependent (y) data

points.
Argument2 is a fact or metric containing numerically independent (x) data

points.

Expression

Usage Notes
l Either Argument1 or Argument2 contains NULL values.
l Argument1 and Argument2 have a different number of values.
Example
See Hypothesis Testing Example, page 83 for an example using the Slope
function.
Standardize
Returns a normalized value from a distribution characterized by a mean and
standard deviation.
Syntax
Standardize(x, Mean, Stdev)
Where:
x is the value you want to normalize.
Stdev is the standard deviation of the distribution.
Expression
Usage Notes
The following is an invalid condition: Stdev £ 0

StandardNormalDistribution (Standard Normal Cumulative

Distribution)
Returns the cumulative distribution of a standard normal variable. The mean
is 0 and the standard deviation is 1. This function can be used in lieu of a
table of standard normal curve areas.
Syntax
StandardNormalDistribution (Argument)
Where:
Argument is a metric representing a list of numbers.
Expression
Usage Notes
This function results in an invalid value if Argument is nonnumeric.
SteYX (Standard Error of Estimates)

Returns the standard error of estimates in linear regression with two
variables (for example, Y = Intercept + Slope * X).
Syntax
SteYX([Y], [X])
Where:
Y is an array or range of dependent data points.
X is an array or range of independent data points.
Expression

Usage Notes
If an argument value contains text, logical values, or empty cells, those

values are ignored; however, cells with the value zero are included.
It is an invalid condition for Y and X to be empty or have a different number

of data points.
TDistribution
Returns the right-tailed distribution of a random variable. This function can
be used in lieu of a table of critical values to test hypotheses on small
sample sets.
Syntax
TDistribution(x, df)
Where:
x is the numeric value at which the distribution is evaluated.
df is a value that indicates the degrees of freedom of the distribution.
Expression
Where:
X is a random value that follows the t-distribution.
Usage Notes
If df is not an integer, it is truncated.

l Either x or df is nonnumeric
l df < 1
Trend
The Trend function uses data recorded for certain time intervals to calculate
and display the expected value of data for an unknown time period. This is
most commonly used to predict future metric values based on historical
data. For example, based on revenue data for three years, you want to

Syntax
Trend <FactID>(KnownData1,...,KnownDataN, FutureTimeOffset)
Where:
KnownData1,...,KnownDataN are separate scalar values that provide the

data to use to predict a future value for. While you can use Trend to enter
these values manually, it is more common to use TrendV which can use a
metric to enter a list of values, as described in TrendV (Trend, Vector Input),
page 488.

FutureTimeOffset.


Example
As described in the syntax requirements, you can use Trend to enter values
manually. However, it is more common to use TrendV to use metrics and
attributes to enter this information, as described in TrendV (Trend, Vector
Input), page 488.
TrendV (Trend, Vector Input)

The TrendV function uses data recorded for certain time intervals to
You can use TrendV to predict values for any point in time, including dates
that have known data. However, this type of function is meant for predicting
future, unknown values.

Syntax
TrendV <FactID, SortBy>(KnownData, FutureTimeOffset)
Where:



FutureTimeOffset.
The TrendV function assumes that data is available for sequential time
intervals with no missing data. If your data meets these requirements, this
allows you to use a single expression to use on reports with varying time
intervals such as year, month, quarter, and so on. This is an advantage of
TrendV in relation to ForecastV, which requires you to specify a time
interval such as year for each expression. For an example, see the examples
described below.


information, see BreakBy and SortBy Parameters, page 27.
Example
The TrendV function is useful to display predicted future values of a metric

as a subtotal of the metric on a report.

2008 Trend can be created with the following definition:
TrendV(x, 1)

metric’s values.
Also, 1 is used as the offset in the time interval to predict data for the first
time interval that is not known.

created. When the 2008 Trend is displayed on the report, it displays the
This value is the same value predicted in the example using ForecastV, as
described in ForecastV (Forecast, Vector Input), page 453. The advantage
of using TrendV is that since you do not have to define the attribute or other
object that defines the time interval, you can switch the time interval used by
modifying the report. For example, the report can be modified to display
Quarter instead of Year, which allows the 2008 Trend subtotal to predict the
revenue for the first quarter of 2008. This is shown in the report below.
The type of subtotal shown above could also be created using ForecastV,
but it would require you to create a separate subtotal that used the Quarter
attribute to define the time intervals.
VarTest (Variance Test)

Returns the P-value of a test that tests the variance of the data against a
particular value.


correct correct
Definition
VarTest returns the P-value for the hypothesis test in the following form:
l H0:
l H1:
Syntax
VarTest <Hypothesis Type> (Argument, Hypothesis Variance)
Where:
Hypothesis Type is the parameter to set whether you are testing the
probability of the value being equal (0), less than (-1), greater than (1), or
not equal to (2) the sigma or variance.
Argument is a fact or metric representing the values you are testing.
HypothesisVariance is the variance you are testing.
Example
VarTest(M1, 1000) = 1000
For a set of values (M1), what is the probability that the variance of the value
will be equal to 1000? Note that M1 needs to have a defined dimensionality.
Usage Notes
For each variance test function, two tests are performed. The first tests the
probability of the value being equal (0), and the second tests the probability
of the value being less than (-1), greater than (1), or not equal to (2) the
sigma or variance. A parameter must be specified for the second test.
WeibullDistribution
Returns either the cumulative distribution or the probability density of a
random variable. Used in reliability analysis, such as the calculation of the
mean time to failure for a given device.
Syntax
WeibullDistribution <Cumulative> (x, Alpha, Beta)
Where:
x is the numeric value at which the distribution is evaluated.

l If Cumulative = 0 (default), it returns the cumulative distribution.
l If Cumulative = 1, it returns the probability density.
Expression
For the Weibull cumulative distribution:
For the Weibull probability density:
When alpha = 1, the Weibull exponential return is
Usage Notes
l alpha or beta is nonnumeric
l x<0
l alpha < 0
l beta < 0

M ICRO STRATEGY AN D
D ATABASE SUPPORT FOR
FUN CTION S

The functions provided with MicroStrategy can be evaluated by the

MicroStrategy Analytical Engine or passed to the database for processing.
The sections and tables in this appendix list MicroStrategy Analytical Engine
and database support for MicroStrategy functions. Reviewing this support
allows you to determine whether a MicroStrategy function can be supported
for your MicroStrategy environment.
l Analytical Engine Support for Functions, page 495
l Databases That a Function Can Be Evaluated On, page 500
For additional information on how functions are processed, see Types of

Function Processing, page 60.
Analytical Engine Support for Functions

The functions listed in the following table are supported by the
MicroStrategy Analytical Engine. This allows metrics to be evaluated by
MicroStrategy in cases where functions cannot be evaluated by a database.
It also allows metrics to support smart metric functionality (see the
Advanced Reporting Help).
Functions supported by the MicroStrategy

Function type
Analytical Engine
• Add
• Average
• Avg
• Count
Basic functions
• First
• GeoMean
• Greatest
• Last


Function type
Analytical Engine
• Least
• Max
• Median
• Min
• Mode
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
All date and time functions are supported by the

Date and time functions
MicroStrategy Analytical Engine
• Banding
• BandingC
Internal functions • BandingP
• Case
• CaseV
• IsNotNull
• IsNull
Null and Zero functions • NullToEmpty
• NullToZero
• ZeroToNull
OLAP functions • ExpWghMovingAvg


Function type
Analytical Engine
• ExpWghRunningAvg
• FirstInRange
• Lag
• LastInRange
• Lead
• MovingAvg
• MovingCount
• MovingDifference
• MovingMax
• MovingMin
• MovingStdev
• MovingStdevP
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum


Function type
Analytical Engine
• NTile
• NTileSize
• NTileValue
Rank and NTile functions
• NTileValueSize
• Percentile
• Rank
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
String functions • LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• All Arithmetic functions are supported by the

Arithmetic operators
MicroStrategy Analytical Engine
• <
• <=
Comparison operators
• <>
• =


Function type
Analytical Engine
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
Comparison operators for rank
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or


Function type
Analytical Engine
All data mining functions are supported by the

Data mining functions
MicroStrategy Analytical Engine.
All financial functions are supported by the

Financial functions
All mathematical functions are supported by the

Mathematical functions
All statistical functions are supported by the

Statistical functions
Databases That a Function Can Be Evaluated On

The tables below list the function support for each database that is certified
for use with MicroStrategy. If a function is listed for a database, the function
can be evaluated within that database. If a function cannot be evaluated
within a database, the function may be able to be supported by the
MicroStrategy Analytical Engine (see Analytical Engine Support for
Functions, page 495).
Except where explicitly stated, only databases that are certified to work with
MicroStrategy 10 are listed below. For information on the latest certification
and support for databases, see Platform Certifications.
Actian Vectorwise
The tables listed below describe the MicroStrategy function support for
Actian Vectorwise databases:
l Actian Vectorwise 2.5, page 501
l Actian Vectorwise 3.0, page 506

Actian Vectorwise 2.5
Function type Functions that can be evaluated on the database
• Add
• Average
• Avg
• Count
• GeoMean
• Greatest
• Least
• Max
Basic functions
• Min
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek

• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
Internal functions
• BandingC
• IsNotNull
• IsNull
Null and Zero functions
• NullToZero
• ZeroToNull
OLAP functions • OLAPRank
Rank and NTile functions • Rank
• Concat
String functions
• ConcatBlank

• InitCap
• LeftStr
• Length
• Lower
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
Arithmetic operators • +
• /
• U-
• <
• <=
• <>
• =
Comparison operators • >
• >=
• Begins With
• Between
• Contains

• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
Data mining functions None
Financial functions None
• Abs
• Acos
• Acosh
• Asin

• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• RandBetween
• Round
• Round2
• Sin
• Sinh
• Sqrt

• Tan
• Tanh
• Trunc
• Correlation
• Covariance
Statistical functions • Intercept
• RSquare
• Slope
Actian Vectorwise 3.0
• Add
• Average
• Avg
• Count
• GeoMean
• Greatest
Basic functions • Least
• Max
• Min
• Multiply
• Product
• Stdev
• StdevP

• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate

• YearStartDate
• Banding
Internal functions
• BandingC
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
Arithmetic operators • x
• +

• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
Comparison operators for rank • *=
• *>=
• *Between

• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log

• Log10
• Mod
• Power
• Quotient
• Radians
• RandBetween
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
• Correlation
• Covariance
• RSquare
• Slope
Aster Database
The tables listed below describe the MicroStrategy function support for Aster
Databases:
l Aster Database 4.6.x, page 512
l Databases That a Function Can Be Evaluated On, page 500

l Aster Database 5.1.x, page 522
Aster Database 4.6.x
• Add
• Average
• Avg
• Count
• GeoMean
• Greatest
• Least
Basic functions • Max
• Min
• Multiply
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
Date and time functions • CurrentDateTime
• CurrentTime
• Date
• DayOfMonth

• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
Internal functions • Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
• Lead
OLAP functions
• MovingAvg
• MovingCount

• MovingMax
• MovingMin
• MovingSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningSum
• Concat
• InitCap
• LeftStr
• Length
• Lower
• LTrim
String functions
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• +
• /

• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between

• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
Mathematical functions • Cos
• Degrees
• Exp
• Floor
• Int
• Ln
• Log
• Log10
• Mod
• Power

• Quotient
• Radians
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
Statistical functions None
• Add
• Average
• Avg
• Count
• GeoMean
• Greatest
Basic functions
• Least
• Max
• Min
• Multiply
• Stdev
• StdevP

• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate

• YearStartDate
Internal functions None
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
OLAP functions
• MovingSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningSum
• Concat
• InitCap
• LeftStr
String functions
• Length
• Lower
• LTrim

• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between

• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
Mathematical functions • Asinh
• Atan
• Atan2
• Atanh
• Ceiling

• Cos
• Degrees
• Exp
• Floor
• Int
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
• Add
Basic functions
• Average

• Avg
• Count
• GeoMean
• Greatest
• Least
• Max
• Min
• Multiply
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond

• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
• Lead
• MovingAvg
• MovingCount
OLAP functions • MovingMax
• MovingMin
• MovingSum
• RunningAvg
• RunningCount

• RunningMax
• RunningMin
• RunningSum
• Concat
• InitCap
• LeftStr
• Length
• Lower
• LTrim
String functions
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =

• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or

• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Degrees
• Exp
• Floor
• Int
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin

• Sqrt
• Tan
• Trunc
Calpont InfiniDB
Calpont InfiniDB databases:
l Calpont InfiniDB 2.2.x, page 528
l Calpont InfiniDB 3.x, page 533
Calpont InfiniDB 2.2.x
• Add
• Average
• Avg
• Count
• GeoMean
Basic functions • Greatest
• Least
• Max
• Min
• Multiply
• Product

• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week

• Year
• YearEndDate
• YearStartDate
• Banding
Internal functions
• BandingC
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None
Rank and NTile functions None
• ConcatBlank
• Concat
• InitCap
• LeftStr
• Length
• Lower
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
Arithmetic operators • -

• x
• +
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
Comparison operators for rank • *<>
• *=

• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Degrees
• Exp
• Floor
• Int
• Ln
• Log

• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
Calpont InfiniDB 3.x
• Add
• Average
• Avg
• Count
Basic functions
• GeoMean
• Greatest
• Least
• Max

• Min
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
Date and time functions • DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate

• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
Internal functions
• BandingC
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None
• ConcatBlank
• Concat
• InitCap
• LeftStr
• Length
String functions • Lower
• LTrim
• Position
• RightStr
• RTrim
• SubStr

• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like

• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
Mathematical functions • Atan2
• Atanh
• Ceiling
• Cos
• Degrees
• Exp
• Floor

• Int
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
EXASolution
EXASolution databases:
l EXASolution 4.x, page 539

EXASolution 4.x
• Add
• Average
• Avg
• Count
• GeoMean
• Greatest
• Least
• Max
Basic functions
• Median
• Min
• Multiply
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek

• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
• Lead
OLAP functions • MovingAvg
• MovingCount
• MovingMax

• MovingMin
• MovingSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningSum
• Concat
• InitCap
• LeftStr
• Length
• Lower
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
Comparison operators • <

• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
Logical operators
• IF

• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
Mathematical functions • Degrees
• Exp
• Floor
• Int
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians

• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
• Correlation
• Covariance
• RSquare
• Slope
Greenplum
Greenplum databases:
l Greenplum 4.x, page 544
Greenplum 4.x
• Add
Basic functions • Average
• Avg

• Count
• GeoMean
• Greatest
• Least
• Max
• Min
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond

• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Year
• YearEndDate
• YearStartDate
• Banding
Internal functions • BandingC
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None
• Concat
• ConcatBlank
• InitCap
String functions
• LeftStr
• Length
• Lower

• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
Comparison operators • Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With

• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh

• Ceiling
• Cos
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
• Correlation
• Covariance
• RSquare
• Slope

Hadoop Hive
Hadoop Hive databases:
l Hadoop Hive, page 550
Hadoop Hive
• Add
• Average
• Avg
• Count
• First
• GeoMean
• Greatest
• Last
• Max
• Min
• Multiply
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• CurrentDate

• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DaysBetween
• Hour
• Minute
• Month
• Second
• Week
• Year
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None
• Concat
• LeftStr
• Length
• LTrim
• Position
• RightStr

• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With

• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Acos
• Acosh
• Asin
• Asinh
• Atan
Mathematical functions • Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp

• Floor
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Covariance
• Standardize
HP Vertica
The tables listed below describe the MicroStrategy function support for HP
Vertica databases:
l HP Vertica 5.1 , page 555
l HP Vertica 6.x , page 560

HP Vertica 5.1
• Add
• Average
• Avg
• Count
• GeoMean
• Max
• Min
Basic functions
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween

• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Year
• YearEndDate
• YearStartDate
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
• LastInRange
• Lead
• MovingAvg
OLAP functions
• MovingCount
• MovingMax
• MovingMin
• MovingStdev
• MovingSum

• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• LTrim
• Position
• RightStr
• RTrim
• SubStr

• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like

• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atanh
• Ceiling
• Cos
• Degrees
• Exp
• Ln

• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
Statistical functions • Standardize
HP Vertica 6.x
• Add
• Average
• Avg
Basic functions • Count
• GeoMean
• Max
• Min

• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter

• Second
• Year
• YearEndDate
• YearStartDate
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
• LastInRange
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
• MovingStdev
OLAP functions
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount

• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-

• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
Logical operators
• IF

• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Exp
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round

• Round2
• Sin
• Sqrt
• Tan
• Trunc
IBM DB2
The tables listed below describe the MicroStrategy function support for IBM
DB2 databases:
l DB2 V9.5 for Linux, UNIX, and Windows, page 566
l DB2 for i 6.1, page 591
l DB2 for i 7.1, page 596
DB2 V9.5 for Linux, UNIX, and Windows
• Add
• Average
Basic functions
• Avg
• Count

• GeoMean
• Max
• Min
• Multiply
• Product
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween

• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
• Lead
• MovingAvg
• MovingCount
• MovingMax
OLAP functions • MovingMin
• MovingStdev
• MovingStdevP
• MovingSum
• OLAPAvg
• OLAPCount

• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper

• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>

• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int

• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
• Correlation
• Covariance
• RSquare
• Slope

• Add
• Average
• Avg
• Count
• GeoMean
• Max
Basic functions • Min
• Multiply
• Product
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
Date and time functions • CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear

• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
• Lead
OLAP functions
• MovingAvg
• MovingCount

• MovingMax
• MovingMin
• MovingStdev
• MovingStdevP
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
• InitCap
String functions
• LeftStr
• Length
• Lower

• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With

• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh

• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
Statistical functions • Correlation

• Covariance
• Intercept
• RSquare
• Slope
• Add
• Average
• Avg
• Count
• GeoMean
• Max
• Multiply
• Product
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime

• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
Null and Zero functions • IsNull
• NullToZero

• ZeroToNull
• Lag
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
• MovingStdev
• MovingStdevP
• MovingSum
• OLAPAvg
• OLAPCount
OLAP functions
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
String functions • Concat

• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between

• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
Mathematical functions • Acos
• Acosh

• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sinh

• Sqrt
• Tan
• Tanh
• Trunc
• Correlation
• Covariance
• RSquare
• Slope
• Add
• Average
• Avg
• Count
• GeoMean
• Max
Basic functions
• Min
• Multiply
• Product
• StdevP
• Sum
• Var

• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
Date and time functions • MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
Internal functions • Banding

• BandingC
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
• MovingStdev
• MovingStdevP
• MovingSum
OLAP functions • OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin

• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
Comparison operators • <=
• <>

• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or

• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Cosh
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians

• Randbetween
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
• Correlation
• Covariance
• RSquare
• Slope
DB2 for i 6.1
• Add
• Average
• Avg
• GeoMean
• Max
• Min

• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter

• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• LTrim
• Position
• RightStr
• RTrim
• SubStr

• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like

• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp

• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
DB2 for i 7.1
Basic functions • Add

• Average
• Avg
• Count
• GeoMean
• Max
• Min
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond

• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None
• Concat
• ConcatBlank
String functions • InitCap
• LeftStr
• Length

• Lower
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like

• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
Mathematical functions • Asin
• Asinh
• Atan
• Atan2

• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc

IBM Informix
Informix databases:
l IBM Informix IDS 11.5 and Informix Ultimate Edition 11.7, page 602
l IBM Informix Ultimate Edition 12.1, page 606
l IBM Informix XPS 8.x, page 611
IBM Informix IDS 11.5 and Informix Ultimate Edition 11.7
• Add
• Average
• Avg
• Count
• Max
Basic functions
• Min
• Multiply
• Stdev
• Sum
• Var
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date

• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None

• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• LTrim
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >=
• Begins With
• Between
• Contains

• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Asin
• Atan

• Atan2
• Ceiling
• Cos
• Exp
• Floor
• Int
• Int2
• Ln
• Log10
• Mod
• Power
• Quotient
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
IBM Informix Ultimate Edition 12.1
• Add
Basic functions
• Average

• Avg
• Count
• Max
• Min
• Multiply
• Stdev
• Sum
• Var
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate

• Quarter
• Second
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
• LastInRange
• Lead
• OLAPAvg
OLAP functions
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• Concat
String functions
• ConcatBlank

• InitCap
• LeftStr
• Length
• Lower
• LTrim
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
Comparison operators • >=
• Begins With
• Between
• Contains
• Ends With
• In

• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Asin
• Atan
• Atan2
• Ceiling

• Cos
• Exp
• Floor
• Int
• Int2
• Ln
• Log10
• Mod
• Power
• Quotient
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
IBM Informix XPS 8.x
• Add
• Average
Basic functions
• Avg
• Count

• Max
• Min
• Multiply
• Stdev
• Sum
• Var
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
Date and time functions • DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second

• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• LTrim
• RightStr
• RTrim
• SubStr
• Trim
• -
• x

• +
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=

• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Asin
• Atan
• Atan2
• Ceiling
• Cos
• Exp
Mathematical functions • Floor
• Int
• Int2
• Ln
• Log10
• Mod
• Power
• Round
• Round2

• Sin
• Sqrt
• Tan
• Trunc
IBM Netezza
Netezza databases:
l IBM Netezza 5.0.x, page 616
IBM Netezza 5.0.x
• Add
• Average
• Avg
• Count
Basic functions • GeoMean
• Max
• Min
• Multiply
• Product

• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week

• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
• LastInRange
• Lead
• MovingAvg
• MovingCount
• MovingMax
OLAP functions • MovingMin
• MovingStdev
• MovingStdevP
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin

• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
String functions
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
Arithmetic operators • x
• +

• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
Comparison operators for rank • *=
• *>=
• *Between

• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log

• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
• Correlation
• Covariance
• Fisher
• Intercept
• InverseFisher
Statistical functions • Pearson
• RSquare
• Skew
• Slope
• Standardize
• SteYX

IBM Netezza 6.0.x
• Add
• Average
• Avg
• Count
• GeoMean
• Max
• Min
Basic functions
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween

• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
OLAP functions • LastInRange
• Lead
• MovingAvg

• MovingCount
• MovingMax
• MovingMin
• MovingStdev
• MovingStdevP
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
String functions • InitCap
• LeftStr
• Length

• Lower
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With

• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh

• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc

• Covariance
• Fisher
• Intercept
• InverseFisher
• Kurtosis
• Pearson
• RSquare
• Skew
• Slope
• Standardize
• SteYX
IBM Netezza 7.0.x
• Add
• Average
• Avg
• Count
• GeoMean
Basic functions
• Max
• Min
• Multiply
• Product
• Stdev

• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
Date and time functions • Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year

• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
• LastInRange
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
OLAP functions
• MovingStdev
• MovingStdevP
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank

• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
String functions
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• +
• /

• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between

• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10

• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
• Correlation
• Covariance
• Fisher
• Intercept
• InverseFisher
• Kurtosis
• Pearson
• RSquare
• Skew
• Slope
• Standardize
• SteYX

IBM Red Brick

Red Brick databases:
l IBM Red Brick 6.3, page 636
IBM Red Brick 6.3
• Add
• Average
• Avg
• Count
Basic functions
• Max
• Min
• Multiply
• Sum
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
Date and time functions • Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour

• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
• Lead
• MovingAvg
OLAP functions
• MovingCount
• MovingMax
• MovingMin

• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
String functions
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim

• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>

• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Ceiling
• Exp
• Floor
• Int
• Int2
• Ln
• Mod
• Quotient
• Round
• Round2
• Sqrt

Infobright
Infobright databases:
l Infobright 4.0.x, page 641
Infobright 4.0.x
• Add
• Average
• Avg
• Count
• GeoMean
• Greatest
• Least
• Max
Basic functions
• Min
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
Date and time functions • AddMonths
• CurrentDate

• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• NullToZero

• ZeroToNull
OLAP functions None
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =

• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or

• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween

• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
Kognitio
Kognitio databases:
l Kognitio WX2 7.x, page 646
Kognitio WX2 7.x
• Add
• Average
• Avg
• First
• GeoMean
• Greatest

• Last
• Least
• Max
• Min
• Multiply
• StdevP
• Sum
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DaysBetween
• Hour
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter

• Second
• Week
• Year
• YearEndDate
• YearStartDate
• NullToZero
• ZeroToNull
• Lag
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
• MovingStdevP
• MovingSum
OLAP functions • OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax

• RunningMin
• RunningStdevP
• RunningSum
• Concat
• InitCap
• LeftStr
• Length
• Lower
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
Comparison operators • <>
• =
• >

• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or

• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
Mathematical functions • Floor
• Int
• Ln
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin
• Sinh
• Sqrt

• Tan
• Tanh
• Trunc
Maria DB
Maria DB databases:
l Maria DB 5.5.x, page 652
Maria DB 5.5.x
• Add
• Average
• Avg
• Count
• GeoMean
• Greatest
Basic functions
• Least
• Max
• Min
• Multiply
• Product
• Stdev

• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate

• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x

• +
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=

• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
Mathematical functions • Ceiling
• Cos
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log

• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
Microsoft Access
Microsoft Access databases:
l Microsoft Access 2000, 2002, or 2003, page 657
l Microsoft Access 2007, page 661
Microsoft Access 2000, 2002, or 2003

Microsoft Access 2000, 2002, or 2003 are only supported for demonstration
purposes.

• Add
• Average
• Avg
• Count
Basic functions
• Max
• Min
• Multiply
• Sum
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DaysBetween
• Hour
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter

• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Position
• RightStr
• SubStr
• Trim
• Upper

• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>

• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Floor
• Int
• Int2
• Ln
• Mod
• Quotient
Microsoft Access 2007

Microsoft Access 2007 is only supported for demonstration purposes.
• Add
Basic functions
• Average

• Avg
• Count
• GeoMean
• Max
• Min
• Multiply
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• Millisecond
• Minute
• Month

• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None
• Concat
• ConcatBlank
• InitCap
• LeftStr
String functions
• Length
• Lower
• LTrim
• Position

• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
Comparison operators • Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains

• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Cos
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod

• Power
• Quotient
• Round
• Round2
• Sin
• Sqrt
• Tan
Microsoft SQL Server

Microsoft SQL Server databases:
l Microsoft SQL Server 2005, page 666
l Microsoft SQL Server 2008 R2 Parallel Data Warehouse, page 677
l Microsoft SQL Server 2012 Parallel Data Warehouse, page 687
l Microsoft SQL Database, page 692
Microsoft SQL Server 2005
• Add
Basic functions
• Average

• Avg
• Count
• GeoMean
• Max
• Min
• Multiply
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month

• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Concat
• ConcatBlank
• InitCap
String functions • LeftStr
• Length
• Lower
• LTrim

• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between

• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
Mathematical functions • Asinh
• Atan
• Atan2
• Atanh
• Ceiling

• Cos
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc

• Add
• Average
• Avg
• Count
• GeoMean
• Max
• Multiply
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour

• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Concat
• ConcatBlank
String functions
• InitCap
• LeftStr

• Length
• Lower
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• Begins With
• Between
• Contains
• Ends With
• In

• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan

• Atan2
• Atanh
• Ceiling
• Cos
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc

Microsoft SQL Server 2008 R2 Parallel Data Warehouse
• Add
• Average
• Avg
• Count
• GeoMean
• Max
• Multiply
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour

• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
OLAP functions • Lead
• OLAPRank
• Concat
String functions
• ConcatBlank

• InitCap
• LeftStr
• Length
• Lower
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >=
• Begins With
• Between
• Contains

• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acosh
• Asin
• Asinh

• Atan
• Atanh
• Ceiling
• Cos
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc

• Add
• Average
• Avg
• Count
• GeoMean
• Max
• Multiply
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour

• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
• LastInRange
• Lead
OLAP functions
• MovingAvg
• MovingCount
• MovingMax

• MovingMin
• MovingStdev
• MovingStdevP
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
• InitCap
• Length
• Lower
• LTrim

• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains

• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atanh
• Atan2
• Ceiling
• Cos

• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
Microsoft SQL Server 2012 Parallel Data Warehouse
• Add
• Average
Basic functions
• Avg
• Count

• GeoMean
• Max
• Min
• Multiply
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween

• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
• OLAPRank
• Concat
• ConcatBlank
• InitCap
• Length
• Lower
• LTrim

• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between

• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acosh
• Asin
• Asinh
Mathematical functions • Atan
• Atanh
• Ceiling
• Cos
• Exp

• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
Microsoft SQL Database
• Add
• Average
Basic functions • Avg
• Count
• GeoMean

• Max
• Min
• Multiply
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate

• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
String functions
• Lower
• LTrim
• Position
• RightStr
• RTrim

• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
Comparison operators • Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In

• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Degrees
• Exp

• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
MySQL
MySQL databases:
l MySQL 5.x, page 698

MySQL 5.x
• Add
• Average
• Avg
• Count
• GeoMean
• Greatest
• Least
• Max
Basic functions
• Min
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek

• DayOfYear
• DaysBetween
• Hour
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None
• Concat
String functions
• ConcatBlank

• InitCap
• LeftStr
• Length
• Lower
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >=
• Begins With
• Between
• Contains

• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin

• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc

Oracle
Oracle databases:
l Oracle 10g, page 703
l Oracle 10gR2, page 710
l Oracle 11g, page 716
l Oracle 11g R2, page 723
l Oracle 12c, page 730
Oracle 10g
• Add
• Average
• Avg
• Count
• First
• GeoMean
• Greatest
Basic functions
• Last
• Least
• Max
• Median
• Min
• Multiply
• Product

• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year

• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
• LastInRange
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
OLAP functions
• MovingStdev
• MovingStdevP
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank

• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
String functions
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• +
• /

• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between

• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10

• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
• Correlation
• Covariance
• Fisher
• Intercept
• InverseFisher
• Kurtosis
• Pearson
• RSquare
• Skew
• Slope
• Standardize
• SteYX

Oracle 10gR2
• Add
• Average
• Avg
• Count
• First
• GeoMean
• Greatest
• Last
• Least
• Median
• Min
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
Date and time functions • CurrentDate
• CurrentDateTime
• CurrentTime

• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions • FirstInRange

• Lag
• LastInRange
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
• MovingStdev
• MovingStdevP
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum

• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >=
• Begins With
• Between
• Contains

• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin

• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan

• Tanh
• Trunc
• Correlation
• Covariance
• Fisher
• Intercept
• InverseFisher
• Kurtosis
• Pearson
• RSquare
• Skew
• Slope
• Standardize
• SteYX
Oracle 11g
• Add
• Average
• Avg
• First
• GeoMean
• Greatest

• Last
• Least
• Max
• Median
• Min
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
Date and time functions • DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• Minute
• Month

• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
• LastInRange
• Lead
OLAP functions • MovingAvg
• MovingCount
• MovingMax
• MovingMin
• MovingStdev

• MovingStdevP
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
String functions • Length
• Lower
• LTrim
• Position
• RightStr

• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
Comparison operators • Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In

• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees

• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
• Correlation
• Covariance
• Fisher
• Intercept
• InverseFisher
• Kurtosis

• Pearson
• RSquare
• Skew
• Slope
• Standardize
• SteYX
Oracle 11g R2
• Add
• Average
• Avg
• Count
• First
• GeoMean
• Greatest
Basic functions • Last
• Least
• Max
• Median
• Min
• Multiply
• Product
• Stdev

• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate

• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
• LastInRange
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
OLAP functions • MovingStdev
• MovingStdevP
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum

• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
String functions
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-

• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
Logical operators • AND

• IF
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod

• Power
• Quotient
• Radians
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
• Correlation
• Covariance
• Fisher
• Intercept
• InverseFisher
• Kurtosis
• Pearson
• RSquare
• Skew
• Slope
• Standardize
• SteYX

Oracle 12c
• Add
• Average
• Avg
• Count
• First
• GeoMean
• Greatest
• Last
• Least
• Median
• Min
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
Date and time functions • CurrentDate
• CurrentDateTime
• CurrentTime

• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions • FirstInRange

• Lag
• LastInRange
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
• MovingStdev
• MovingStdevP
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum

• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >=
• Begins With
• Between
• Contains

• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin

• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan

• Tanh
• Trunc
• Correlation
• Covariance
• Fisher
• Intercept
• InverseFisher
• Kurtosis
• Pearson
• RSquare
• Skew
• Slope
• Standardize
• SteYX
Actian Matrix
Actian Matrix databases:
Actian Matrix was previously referred to as ParAccel.
l Actian Matrix 3.1.x, page 737

Actian Matrix 3.1.x
• Add
• Average
• Avg
• Count
• GeoMean
• Min
• Multiply
• Product
• Sum
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate

• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
• LastInRange
• Lead
• MovingAvg
OLAP functions
• MovingCount
• MovingMax
• MovingMin
• MovingSum
• OLAPAvg

• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
String functions
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x

• +
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=

• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Ln
• Log

• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
• Covariance
• Fisher
• Intercept
• InverseFisher
• Kurtosis
• RSquare
• Slope
• Standardize

Actian Matrix 3.5.x
• Add
• Average
• Avg
• Count
• GeoMean
• Max
• Min
Basic functions
• Multiply
• Product
• Sum
• Stdev
• StdevP
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DaysBetween
• Hour
• MilliSecond

• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
• LastInRange
• Lead
OLAP functions
• MovingAvg
• MovingCount
• MovingMax
• MovingMin

• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
String functions
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim

• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>

• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int

• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sinh
• Tan
• Tanh
• Trunc
• Covariance
• Fisher
• Intercept
• InverseFisher
• Kurtosis
• RSquare
• Slope
• Standardize

Actian Matrix 4.0.x
• Add
• Average
• Avg
• Count
• GeoMean
• Max
• Min
Basic functions
• Multiply
• Product
• Sum
• Stdev
• StdevP
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DaysBetween
• Hour
• MilliSecond

• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
• LastInRange
• Lead
OLAP functions
• MovingAvg
• MovingCount
• MovingMax
• MovingMin

• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
String functions
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim

• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>

• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int

• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sinh
• Tan
• Tanh
• Trunc
• Covariance
• Fisher
• Intercept
• InverseFisher
• Kurtosis
• RSquare
• Slope
• Standardize

PostgreSQL
PostgreSQL databases:
l PostgreSQL 8.4, page 755
l PostgreSQL 9.x, page 760
PostgreSQL 8.4
• Add
• Average
• Avg
• Count
• GeoMean
• Greatest
• Least
• Max
Basic functions
• Min
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths

• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull

• NullToZero
• ZeroToNull
• FirstInRange
• OLAPRank
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-

• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
Logical operators
• IF

• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient

• Radians
• Randbetween
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
• Correlation
• Covariance
• RSquare
• Slope
PostgreSQL 9.x
• Add
• Average
• Avg
• Count
Basic functions
• GeoMean
• Greatest
• Least
• Max

• Min
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate

• Quarter
• Second
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• FirstInRange
• OLAPRank
• Concat
• ConcatBlank
• InitCap
• LeftStr
String functions • Length
• Lower
• LTrim
• Position
• RightStr

• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With

• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
Mathematical functions • Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Degrees

• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
• Correlation
• Covariance
• RSquare
• Slope

Salesforce.com
Salesforce.com data sources:
l Salesforce.com, page 766
Salesforce.com
• Add
• Average
• Avg
• Count
• Max
• Min
• Multiply
• Sum
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond

• Minute
• Month
• MonthsBetween
• Quarter
• Second
• Week
• Year
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• LTrim
• Position
• RightStr
• RTrim
• SubStr

• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like

• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Ceiling
• Cos
• Cosh
• Degrees
• Exp

• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Randbetween
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
SAND CDBMS
SAND CDBMS databases:
l SAND CDBMS 6.1, page 771
l SAND CDBMS 7.1, page 775

SAND CDBMS 6.1
• Add
• Average
• Avg
• Count
• GeoMean
• Max
Basic functions • Median
• Min
• Multiply
• Product
• Stdev
• Sum
• Var
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour

• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None
• Concat
• ConcatBlank
• InitCap
String functions
• LeftStr
• Length
• Lower

• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With

• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atanh
• Ceiling

• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Ln
• Log10
• Power
• Radians
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Correlation
Statistical functions • Covariance
• Intercept
SAND CDBMS 7.1

• Average
• Avg
• Count
• GeoMean
• Max
• Median
• Min
• Multiply
• Product
• Stdev
• Sum
• Var
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute

• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None
• Concat
• ConcatBlank
• InitCap
• LeftStr
String functions
• Length
• Lower
• LTrim
• Position

• RightStr
• RTrim
• SubStr
• Trim
• Upper
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains

• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atanh
• Ceiling
• Cos
• Cosh

• Degrees
• Exp
• Floor
• Int
• Ln
• Log10
• Power
• Radians
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Correlation
Statistical functions • Covariance
• Intercept
SAP HANA 1.x

The tables listed below describe the MicroStrategy function support for SAP
HANA 1.x databases:
l SAP HANA 1.0 SP4 , page 781
l SAP HANA 1.0 SP5 , page 785

SAP HANA 1.0 SP4
• Add
• Average
• Avg
• Count
• GeoMean
• Greatest
• Max
• Min
• Multiply
• Stdev
• Sum
• Var
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour

• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
Internal functions
• BandingC
• NullToZero
• ZeroToNull
OLAP functions None
String functions None
• -
• x
• /
• U-

• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• NotIn
• NotLike
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
Logical operators
• IF

• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
Mathematical functions • Cosh
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient

• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
SAP HANA 1.0 SP5
• Add
• Average
• Avg
• Count
• GeoMean
• Greatest
Basic functions
• Least
• Max
• Min
• Multiply
• Stdev
• Sum

• Var
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
Internal functions • Banding

• BandingC
• NullToZero
• ZeroToNull
• Lag
• OLAPRank
String functions None
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like

• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• NotIn
• NotLike
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
Mathematical functions • Asin
• Asinh
• Atan
• Atan2

• Atanh
• Ceiling
• Cos
• Cosh
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc

Sybase
The tables listed below describe the MicroStrategy function support for SAP
Sybase databases:
l SAP Sybase ASE 15.x, page 790
l SAP Sybase IQ 15.2, page 794
SAP Sybase ASE 15.x
• Add
• Average
• Avg
• Count
Basic functions
• Max
• Min
• Multiply
• Sum
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date

• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
OLAP functions None

• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >=
• Begins With
• Between
• Contains

• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Asin
• Atan

• Atan2
• Ceiling
• Cos
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin
• Sqrt
• Tan
SAP Sybase IQ 15.2

• Average
• Avg
• Count
• GeoMean
• Max
• Median
• Min
• Multiply
• Stdev
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute

• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
• LastInRange
• Lead
OLAP functions
• MovingAvg
• MovingCount
• MovingMax
• MovingMin

• MovingStdev
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
String functions
• Lower
• LTrim
• Position
• RightStr
• RTrim

• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like

• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Asin
• Atan
• Atan2
• Ceiling
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln

• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
SAP Sybase IQ 15.3
• Add
• Average
• Avg
• Count
• Max
• Median
• Min
• Multiply

• Stdev
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year

• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
• LastInRange
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
OLAP functions
• MovingStdev
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum

• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
String functions
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-

• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
Logical operators
• IF

• Not
• Or
• Abs
• Acos
• Asin
• Atan
• Atan2
• Ceiling
• Cos
• Degrees
• Exp
• Floor
Mathematical functions • Int
• Int2
• Ln
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin

• Sqrt
• Tan
• Trunc
SAP Sybase IQ 15.4
• Add
• Average
• Avg
• Count
• GeoMean
• Max
• Min
• Multiply
• Stdev
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime

• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• NullToZero

• ZeroToNull
• FirstInRange
• Lag
• LastInRange
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
• MovingStdev
• MovingSum
• OLAPAvg
OLAP functions
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningSum

• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >=
• Begins With
• Between
• Contains

• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Asin
• Atan

• Atan2
• Ceiling
• Cos
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc

SAP Sybase IQ 16.0
• Add
• Average
• Avg
• Count
• GeoMean
• Max
• Min
• Multiply
• Stdev
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour

• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• FirstInRange
• Lag
• LastInRange
OLAP functions
• Lead
• MovingAvg
• MovingCount

• MovingMax
• MovingMin
• MovingStdev
• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
String functions
• Length
• Lower
• LTrim
• Position

• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With

• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Asin
• Atan
• Atan2
• Cos
• Degrees
• Exp
• Floor
• Int

• Int2
• Ln
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin
• Sqrt
• Tan
• Trunc
Teradata
Teradata databases:
l Teradata 12, page 818
l Teradata 13, page 824
l Teradata 13.10, page 830

Teradata 12
• Add
• Average
• Avg
• Count
• GeoMean
• Max
• Min
Basic functions
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfYear
• DaysBetween
• Hour

• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
• Lead
• MovingAvg
OLAP functions • MovingCount
• MovingMax
• MovingMin
• MovingStdev

• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
String functions
• Lower
• LTrim
• Position
• RightStr
• RTrim

• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like

• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp

• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
• Correlation
• Covariance
• Fisher
• Intercept
• InverseFisher
• Kurtosis
• Pearson
• RSquare

• Skew
• Slope
• Standardize
• SteYX
Teradata 13
• Add
• Average
• Avg
• Count
• GeoMean
• Max
• Min
Basic functions
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
Date and time functions • AddMonths
• CurrentDate

• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfYear
• DaysBetween
• Hour
• MilliSecond
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull

• Lag
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
• MovingStdev
• MovingSum
• OLAPAvg
• OLAPCount
OLAP functions • OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
• Concat
String functions • ConcatBlank
• InitCap

• LeftStr
• Length
• Lower
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• Begins With
• Between
• Contains
• Ends With
• In

• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan

• Atan2
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc

• Covariance
• Fisher
• Intercept
• InverseFisher
• Kurtosis
• Pearson
• RSquare
• Skew
• Slope
• Standardize
• SteYX
Teradata 13.10
• Add
• Average
• Avg
• Count
• GeoMean
Basic functions
• Max
• Min
• Multiply
• Product
• Stdev

• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfYear
• DaysBetween
• Hour
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Year
• YearEndDate
• YearStartDate

• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
• MovingStdev
• MovingSum
• OLAPAvg
OLAP functions
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin

• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
String functions
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =

• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or

• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Round

• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
• Correlation
• Covariance
• Fisher
• Intercept
• InverseFisher
• Kurtosis
• Pearson
• RSquare
• Skew
• Slope
• Standardize
• SteYX
Teradata 14.0
• Add
Basic functions • Average
• Avg

• Count
• GeoMean
• Greatest
• Least
• Max
• Min
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP
• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• MilliSecond

• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
• Lead
• MovingAvg
OLAP functions • MovingCount
• MovingMax
• MovingMin
• MovingStdev

• MovingSum
• OLAPAvg
• OLAPCount
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP
• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
String functions
• Lower
• LTrim
• Position
• RightStr
• RTrim

• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=
• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like

• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
• Abs
• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp

• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin
• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
• Correlation
• Covariance
• Fisher
• InverseFisher
• Kurtosis
• Pearson

• RSquare
• Skew
• Slope
• Standardize
• SteYX
Teradata 14.10
• Add
• Average
• Avg
• Count
• GeoMean
• Greatest
• Least
• Max
Basic functions
• Min
• Multiply
• Product
• Stdev
• StdevP
• Sum
• Var
• VarP

• AddDays
• AddMonths
• CurrentDate
• CurrentDateTime
• CurrentTime
• Date
• DayOfMonth
• DayOfWeek
• DayOfYear
• DaysBetween
• Hour
• Minute
• Month
• MonthEndDate
• MonthsBetween
• MonthStartDate
• Quarter
• Second
• Week
• Year
• YearEndDate
• YearStartDate
• Banding
Internal functions
• BandingC

• Coalesce
• IsNotNull
• IsNull
• NullToZero
• ZeroToNull
• Lag
• Lead
• MovingAvg
• MovingCount
• MovingMax
• MovingMin
• MovingStdev
• MovingSum
• OLAPAvg
• OLAPCount
OLAP functions
• OLAPMax
• OLAPMin
• OLAPRank
• OLAPSum
• RunningAvg
• RunningCount
• RunningMax
• RunningMin
• RunningStdev
• RunningStdevP

• RunningSum
• Concat
• ConcatBlank
• InitCap
• LeftStr
• Length
• Lower
String functions
• LTrim
• Position
• RightStr
• RTrim
• SubStr
• Trim
• -
• x
• /
• U-
• <
• <=
• <>
• =
• >
• >=

• Begins With
• Between
• Contains
• Ends With
• In
• Like
• Not Begins With
• Not Between
• Not Contains
• Not Ends With
• Not In
• Not Like
• *<=
• *<>
• *=
• *>=
• *Between
• Not*Between
• AND
• IF
Logical operators
• Not
• Or
Mathematical functions • Abs

• Acos
• Acosh
• Asin
• Asinh
• Atan
• Atan2
• Atanh
• Ceiling
• Cos
• Cosh
• Degrees
• Exp
• Floor
• Int
• Int2
• Ln
• Log
• Log10
• Mod
• Power
• Quotient
• Radians
• Round
• Round2
• Sin

• Sinh
• Sqrt
• Tan
• Tanh
• Trunc
• Correlation
• Covariance
• Fisher
• Intercept
• InverseFisher
• Kurtosis
• Pearson
• RSquare
• Skew
• Slope
• Standardize
• SteYX

MicroStrategy Functions Reference

Uploaded by

Copyright:

Available Formats

MicroStrategy Functions Reference

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MicroStrategy Functions Reference

Uploaded by

Copyright:

Available Formats

Fu n ct i on s

The Basics of MicroStrategy Objects, Object Definitions, and Functions 6

Additional Examples of Functions in Expressions 83

2. Standard Functions 114

Basic Functions 116

Arithmetic Operators 320

4. Plug-In Package Functions 342

Data Mining Functions 343

A. MicroStrategy and Database Support for Functions 494

Analytical Engine Support for Functions 495

Copyright © 2023 All Rights Reserved 5

This chapter provides the following information:

l The Basics of MicroStrategy Objects, Object Definitions, and Functions,

l Function Syntax and Formula Components, page 9

l Using Functions in Expressions, page 35

l Adding Functions to Expressions with the Insert Function Wizard, page 57

l How MicroStrategy Processes Functions, page 60

l Additional Examples of Functions in Expressions, page 83

The Basics of MicroStrategy Objects, Object

An object is a basic building block in MicroStrategy. There are three types of

MicroStrategy objects are created, maintained and deleted by you. Your

Many objects require you to specify an expression when creating or

Copyright © 2023 All Rights Reserved 6

l New York AND Los Angeles

Expressions can be used in many places. Metrics in particular often require

Copyright © 2023 All Rights Reserved 7

various Profit entries in a data warehouse to arrive at one final number to

Functions Commonly Used to Create Specific Objects

Function MicroStrategy Objects That Can Use

Copyright © 2023 All Rights Reserved 8

Function MicroStrategy Objects That Can Use

• ApplyAgg • Custom group

Apply • ApplyOLAP • Fact

Function Syntax and Formula Components

Each function is designated by a function name. Functions operate on an

Copyright © 2023 All Rights Reserved 9

l Single-value functions (also known as Non-grouping or Scalar

l Group-value functions (also known as Grouping or Aggregate functions):

l OLAP functions (also known as Relative functions): These functions take

Copyright © 2023 All Rights Reserved 10

MovingAvg, NTile, ApplyOLAP, and so on. For details, see OLAP

l Comparison operators: These operators compare single values or lists of

l Logical operators: These operators provide basic comparisons and

l Apply functions: These functions provide access to functions and

In this guide, lists of attributes, metrics, values, parameters, etc. within a

Single-value functions can be used to create facts, metrics, attribute forms,

Copyright © 2023 All Rights Reserved 11

For example, using a single-value function that operates on four arguments,

Using the single-value "+" (addition) function, A+B=C

l A=1 (an argument containing one component whose value is 1)

l B=2 (an argument containing a single component whose value is 2)

l C=3 (the returned value, containing a single output component, whose

In the example above, A, B, and C each contain one component. More

Exam ple 1: Transform ed Fact

Copyright © 2023 All Rights Reserved 12

Suppose Account Transactions is a list of the following values: -300.5, -7.7,

Exam ple 2: Com pound Metric

In both of the previous examples, both single-value and group-value

Copyright © 2023 All Rights Reserved 13

Group-value functions can be used to create simple metrics, nested metrics,