R Programming For ANY Team
R Programming For ANY Team
R PROGRAMMING
IS EASY
LINEAR
REGRESSION
Introduction
IS EASY
I am a hands-on analytics professional and experienced leader. Over the years
conducting analyses, supporting various business functions, and advising
executives, I have come to believe that in the future:
LINEAR
Make no mistake, the ability to conduct powerful data analyses that drive business
REGRESSION
results are not the sole domain of "Data Scientists" or "Data Analysts."
In business analytics, the 80/20 Rule applies - 20% of analytics drive 80% of business
return on investment (ROI).
Don't believe it? Keep reading to see how Excel skills makes learning R easy.
-Dave
2
WHY R PROGRAMMING IS EASY
Table of Contents
Excel Users Write Code................................4
Excel Code
R Code
Excel Skill Translate Directly to R
3
WHY R PROGRAMMING IS EASY
While most Excel users don't think of it this way, they spend a lot of time writing and
debugging code in Excel. In fact, Microsoft Excel is by far and away the world's most
popular programming environment.
Take the image below as example. The user is using Excel's AVERAGE function to
calculate the average of a column (i.e., Petal.Width) of a table (i.e., iris_data) in a
worksheet.
LINEAR
Once the user hits the <enter> key, Excel attempts to interpret the instructions in the
cell and perform the desired operation. If Excel doesn't understand what the user
REGRESSION
typed, it reports an error.
That's coding!
4
WHY R PROGRAMMING IS EASY
R Code
While it isn't the only way to code in Excel, calling Excel functions as depicted on the
previous page is by far the most common. When using Excel in this way, Excel is
operating as a code interpreter.
Using R as an interpreter is very common. The R user types some code and hits the
<enter> key. R then tries to interpret the code, throwing an error if doesn't
understand what was typed by the user.
In this way Excel and R are very similar, but it doesn't stop there. Even the code is
very similar!
LINEAR
The image below depicts the same scenario as on the previous page, but using R
instead. REGRESSION
As depicted, the user is calculating the average (mean is just another name
for the average) of the Petal.Width column of the iris_data table.
5
WHY R PROGRAMMING IS EASY
While this example might seem simple, it demonstrates why R is the fastest, easiest
way for ANY team to unlock advanced analytics.
As you will see through the rest of this document, Excel is a powerful analytical tool
with many concepts and skills that need to be mastered to use Excel effectively.
This knowledge makes the learning process an exercise in mapping Excel skills to R.
LINEAR
REGRESSION
6
WHY R PROGRAMMING IS EASY
As depicted below, Excel features above the water line (e.g., Pivot Tables) only
scratch the surface of Excel's capabilities. However, these feature represent the bulk
of Excel's use in practice.
Another similarity between Excel and R is the "choose your own adventure" aspect
of the technologies. Just as many Excel users never learn Power Query, not every R
user needs to learn statistical analysis to be effective in their work.
LINEAR
REGRESSION
Tables Data Frames
Common Functions Common Functions
Pivot Tables dplyr
Charts ggplot2
7
WHY R PROGRAMMING IS EASY
Excel tables can also be thought of as container objects. Tables contain rows, columns,
cells, data formats, etc.
You probably can see where I'm going with this already.
LINEAR
When analyzing data with R, it's all about the tables - just like Excel.
REGRESSION
Once again, your Excel knowledge directly translates to R.
8
WHY R PROGRAMMING IS EASY
The image below demonstrates how Excel tables are objects. For example, every
table in Excel has a name - whether you explicitly name a table or not. Table names
allow you to directly access/manipulate tables using Excel code.
9
WHY R PROGRAMMING IS EASY
Things work in R exactly the same way. Tables of data in R (known as "data frames")
have names just like Excel tables so that you can write R code to access/manipulate
tables of data.
10
WHY R PROGRAMMING IS EASY
Working with cells of data is very common in Excel. It is useful to think of cells as
objects contained within tables - as depicted below. Once again, you use Excel code
to access cells.
11
WHY R PROGRAMMING IS EASY
12
WHY R PROGRAMMING IS EASY
Excel code supports different ways of accessing columns of data within tables. Two
examples:
13
WHY R PROGRAMMING IS EASY
Notice how similar the actual R code is to Excel when using object names.
14
WHY R PROGRAMMING IS EASY
The bulk of code Excel users write call functions. Often, these function calls are nested
and can be difficult to bug (again, that's coding!).
15
WHY R PROGRAMMING IS EASY
The contrived example from the previous page is repeated here using R code.
A IsSetosa column is being added to the iris.data table (or data frame) and populated
with new data derived from the existing Species column.
First, notice how the workflow is exactly the same as in Excel - only everything is
done in code with R.
Second, notice how similar the R ifelse function call is to Excel code.
16
WHY R PROGRAMMING IS EASY
Common Functions
Like Excel, R comes out of the box with many, many functions to work with columns
of data.
Many of the R functions share the same name with the corresponding Excel
function. In other cases, mapping your Excel knowledge to R is straightforward, as is
depicted below.
17
WHY R PROGRAMMING IS EASY
Excel is a great data visualization tool, supporting many ways to analyze data
visually.
18
WHY R PROGRAMMING IS EASY
R Data Visualizations
R also easily produces data visualizations that are difficult, or not possible to do with
out of the box Excel features.
19
WHY R PROGRAMMING IS EASY
R programming
Business analysis
Linear regression
Logistic regression
Machine learning
This is an intensive 3-day transformation designed to take ANY team from analyzing
data using Excel pivot tables to analyzing data using machine learning in R.
This transformation is applicable to any business domain. The skills your team will
learn can be used to answer diverse questions like:
20
WHY R PROGRAMMING IS EASY
21