Lecture 7: Continuous Random Variable: Donglei Du (Ddu@unb - Edu)

Lecture 7: Continuous Random Variable
Donglei Du
(ddu@unb.edu)
Faculty of Business Administration, University of New Brunswick, NB Canada Fredericton

E3B 9Y2
Donglei Du (UNB) ADM 2623: Business Statistics 1 / 53

Table of contents
1 Continuous Random Variable
Probability Density Function (pdf)
Probability of any set of real numbers
2 Normal Random Variable
Standard Normal Random Variable
General Normal Random Variable
3 Relationship between Z ∼ N (0, 1) and X ∼ N (µ, σ 2 )
4 Calculations with Standard Normal Random Variable via the Normal
Table
Given z-value, calculate probability
Given probability, calculate z-value
5 Calculations with General Normal Random Variable via the Normal
Table
Given x-value, calculate probability
Given probability, calculate x-value
Layout
Table
Table
Continuous Random Variable
A continuous random variable is any random variable whose set of all

the possible values is uncountable.

A probability density function (pdf) for any continuous random

variable is a function f (x) that satisfies the following two properties:
(i) f (x) is nonnegative; namely,
f (x) ≥ 0
(ii) The total area under the curve defined by f (x) is 1; namely
Z ∞
f (x)dx = 1
−∞

Given a continuous random variable X with its probability density

function f (x), for any set B of real numbers, the probability of B is
given by Z
P (X ∈ B) = f (x)dx
B
For instance, if B = [a, b], then the probability of B is given by
Z b
P (a ≤ X ≤ b) = f (x)dx
a
Geometrically, the probability of B is the area under the curve f (x).

Example
Consider the continuous random variable X with its probability

density function f (x) defined below
(
2x, 0 ≤ x ≤ 1
f (x) =
0, x>1
For instance, the probability of [1/3, 2/3] is given by

Z 2/3
P (1/3 ≤ X ≤ 2/3) = 2xdx = (2/3)2 − (1/3)2 = 1/3.
1/3
Geometrically, the probability of [1/3, 2/3] is the area under the curve
f (x) between [1/3, 2/3] .

Example
f (x)
1 2 x
3 3 1

Note
When dealing with a continuous random variable, we assume that the

probability that the variable will take on any particular value is 0!
Instead, probabilities are assigned to intervals of values!
Therefore, give a continuous random variable X, then for any
constant a:
P (X = a) = 0

Layout
Table
Table
The standard normal random variable Z has the following probability

density function:
1 1 2
φ(z) = √ e− 2 z .
2π
0.4
0.3
0.2
y
0.1
0.0
−4 −2 0 2 4

Standard normal curve: Plot R code
> x<-seq(-4,4,length=200)
>y<-dnorm(x,mean=0,sd=1)
> plot(x,y,type="l",lwd=2,col="red")

Properties of Standard Normal Random Variable
The pdf is symmetric around its mean x = 0, which is at the same

time the mode, the median of the distribution.
It is unimodal.
It has inflection points at +1 and -1.
Z has zero mean and unit variance; namely
E[Z] = 0
V[Z] = 1

A general normal distribution has the following probability density

function for any given parameters µ and σ ≥ 0:
1 1 x−µ 2
f (x) = √ e− 2 ( σ ) .
2πσ
The normal distribution is also often denoted as
X ∼ N (µ, σ 2 ).

General Normal Random Variable: µ = 10 and σ = 2
0.20
0.15
0.10
y
0.05
0.00
4 6 8 10 12 14 16

General normal curve: Plot R code
> mu<-10
>sigma<-2
>x<-seq(mu-3*sigma,mu+3*sigma,length=200)
>y<-dnorm(x,mean=mu,sd=sigma)
>plot(x,y,type="l",lwd=2,col="red")

Properties of Standard Normal Random Variable
The pdf is symmetric around its mean x = µ, which is at the same

time the mode, the median of the distribution.
It is unimodal.
It has inflection points at µ ± σ.
X has mean µ and variance σ; namely
E[Z] = µ
V[Z] = σ
The 68-95-99.7 (empirical) rule, or the 3-sigma rule: About 68%

of values drawn from a normal distribution are within one standard
deviation away from the mean; about 95% of the values lie within
two standard deviations; and about 99.7% are within three standard
deviations.

Layout
Table
Table
Relationship between Z ∼ N (0, 1) and X ∼ N (µ, σ 2 )
Given X ∼ N (µ, σ 2 ), then
X −µ
Z= ∼ N (0, 1)
σ
Given Z ∼ N (0, 1), then
X = µ + σZ ∼ N (µ, σ 2 )

Layout
Table
Table
The normal table

Example: Calculate the area between 0 and 1.23.

Solution: The area is equal to the probability between 0 and 1.23
under the standard normal curve. So from the table
P (0 ≤ Z ≤ 1.23) = 0.3907.
0.4
0.3
0.2
y
0.1
0.0
−4 −2 0 2 4
x
R code
> mu<-1
>sigma<-0
> pnorm(1.23, mean=mu, sd=sigma)-0.5
#[1] 0.3906514
Or simply run the following code for the standard normal distribution
where µ = 0 and σ = 1
> pnorm(1.23)-0.5
#[1] 0.3906514

Example: Calculate the area between -2.15 and 2.23.
Solution: The area is equal to the probability between -2.15 and 1.23
under the standard normal curve. So from the table
P (−2.15 ≤ Z ≤ 2.23) = P (−2.15 ≤ Z ≤ 0) + P (0 ≤ Z ≤ 2.23)
= P (0 ≤ Z ≤ 2.15) + P (0 ≤ Z ≤ 2.23)
0.4
0.3
0.2 = 0.4842 + 0.4871 = 0.9713
y
0.1
0.0
−4 −2 0 2 4
R code
> z1<--2.15
> z2<-2.23
> mu<-0
> sigma<-1
> pnorm(z2, mean=mu, sd=sigma)-pnorm(z1, mean=mu, sd=sigma)
[1] 0.9713487

Example: Given the area between 0 and z is 0.3264, find z.

Solution: We want to find z such that
P (0 ≤ Z ≤ z) = 0.3264
From the table, we find z = 0.94.

0.4
0.3
0.2
y
0.1
0.0
−4 −2 0 2 4
x
R code
>p<-0.3264
>qnorm(p+0.5)
[1] 0.9400342

Example: Given the area between less than z is 0.95, find z.

Solution: We want to find z such that
P (Z ≤ z) = 0.95 ⇔ P (0 ≤ Z ≤ z) = 0.95 − 0.5 = 0.45
From the table, we find z = 1.65.

0.4
0.3
0.2
y
0.1
0.0
−4 −2 0 2 4
x
R code
>p<-0.95
>qnorm(p)
[1] 1.644854

Layout
Table
Table
Example: Given a normal random variable X ∼ N (50, 82 ), calculate
the area between 50 and 60.
Solution: The area is equal to the probability between 50 and 60
under the normal curve.

50 − 50 X −µ 60 − 50
P (50 ≤ X ≤ 60) = P ≤ ≤
8 σ 8
= P (0 ≤ Z ≤ 1.25) = 0.394
0.4
0.3
0.2
y
0.1
0.0

R code
> pnorm(1.25)-0.5
[1] 0.3943502

Example: Given a normal random variable X ∼ N (50, 82 ), calculate
the area between 40 and 60.
Solution: The area is equal to the probability between 40 and 60
under the normal curve.

40 − 50 X −µ 60 − 50
P (40 ≤ X ≤ 60) = P ≤ ≤
8 σ 8
= P (−1.25 ≤ Z ≤ 1.25) = 2P (0 ≤ Z ≤ 1.25)
= 2(0.394) = 0.688
0.4
0.3
0.2
y
0.1

R code
> pnorm(1.25)-pnorm(-1.25)
[1] 0.7887005

Example: Given a normal random variable X ∼ N (50, 82 ), and the

area below x is 0.853, find x?
Solution: We want to find x such that

X −µ x−µ
P (X ≤ x) = 0.853 ⇔ P ≤ = 0.853
σ σ
⇔ P (Z ≤ z) = 0.853
⇔ P (0 ≤ Z ≤ z) = 0.853 − 0.5 = 0.353,
where
x−µ
z :=
σ
From the table, we find z = 1.05, implying that
x = µ + zσ = 50 + 1.05(8) = 58.4

Plot
0.4
0.3
0.2
y
0.1
0.0
−4 −2 0 2 4

R code
>p<-0.853
>qnorm(p)
[1] 1.049387

Practical example
Example: Professor X has determined that the scores in his statistics

course are approximately normally distributed with a mean of 72 and
a standard deviation of 5. He announces to the class that the top 15
percent of the scores will earn an A.
Problem: What is the lowest score a student can earn and still
receive an A?

Practical example
Solution: Let X be the students’ scores. Then X ∼ N (72, 52 ). Let

x be the score that separates an A from the rest. Then

X −µ x−µ
P (X ≥ x) = 0.15 ⇔ P ≥ = 0.15
σ σ
⇔ P (Z ≥ z) = 0.15
⇔ P (0 ≤ Z ≤ z) = 0.5 − 0.15 = 0.35,
where
x−µ
z :=
σ
From the table, we find z = 1.04, implying that
x = µ + zσ = 72 + 1.04(5) = 77.2

Plot
0.4
0.3
0.2
y
0.1
0.0
−4 −2 0 2 4

R code
>p<-0.85
>qnorm(p)
[1] 1.036433

Practical example
Example: A manufacturer of aircraft is likely to be very concerned

about the ability of potential users to use the product. If a lot of
pilots cannot reach the rudder pedals or the navigation systems, then
there is trouble. Suppose a manufacturer knows that the lengths of
pilot’s legs are normally distributed with mean 76 and standard
deviation of 5 cm.
Problem: If the manufacturer wants to design a cockpit such that
precisely 90% of pilots can reach the rudder pedals with their feet
while seated, what is the desired distance between seat and pedals?

Practical example
Solution: Let X be the the lengths of pilot’s legs. Then

X ∼ N (76, 52 ). Let x be the desired distance. Then

X −µ x−µ
P (X ≥ x) = 0.90 ⇔ P ≥ = 0.90
σ σ
⇔ P (Z ≥ z) = 0.90
⇔ P (z ≤ Z ≤ 0) = 0.9 − 0.5 = 0.4
⇔ P (0 ≤ Z ≤ −z) = 0.4
where
x−µ
z :=
σ
From the table, we find z = −1.28, implying that
x = µ + zσ = 76 − 1.28(5) = 69.6

Plot
0.4
0.3
0.2
y
0.1
0.0
−4 −2 0 2 4

R code
>p<-0.10
>qnorm(p)
[1] -1.281552

Practical example
Example: Suppose that a manufacturer of aircraft engines knows

their lifetimes to be a normally distributed random variable with a
mean of 2,000 hours and a standard deviation of 100 hours
Problem: What is the probability that a randomly chosen engine has
a lifetime between 1,950 and 2,150 hours?

Practical example
Solution: Let X be the lifetimes of their aircraft engines. Then

X ∼ N (2000, 1002 ). Then
P (1950 ≤ X ≤ 2150)

1950 − 2000 X −µ 2150 − 2000
= P ≤ ≤
100 σ 100
= P (−0.5 ≤ Z ≤ 1.5) = P (0 ≤ Z ≤ 0.5) + P (0 ≤ Z ≤ 1.5)
= 0.1915 + 0.4332 = 0.6247.

Plot
0.4
0.3
0.2
y
0.1
0.0
−4 −2 0 2 4

R code
> z1<--0.5
> z2<-1.5
> pnorm(z2)-pnorm(z1)
[1] 0.6246553

The Probability of a market Crash
Example: Suppose that the annualized S&P 500 index returns,

µ ≈ 12% and σ ≈ 15%.
Problem: A negative surprise: on October 19, 1987, the S&P 500
index dropped more than 23% on one day. What is the probability for
such a event?

Solution
Solution: Let r denote the daily return, then r is normally distributed

with
mean
0.12/252 ≈ 0.00048,
and standard deviation
√
0.15/ 252 = 0.0094.
Namely r ∼ N (0.00048, 0.00942 ). Then
P (r ≤ −0.23)

r−µ −0.23 − 0.00048
= P ≤
σ 0.0094
= P (Z ≤ −24) ≈ 10−127 .

The empirical rule
We now derive the empirical rule (Back in Lecture 4) from the

Normal table: assume X ∼ N (µ, σ 2 ), then

X −µ
P (µ − kσ ≤ X ≤ µ − kσ) = P −k ≤ ≤ k = P(−k ≤ Z ≤ k)
σ
For k = 1, 2, 3, we obtain 0.68, 0.95, and 99.7 from the Normal table.

Lecture 7: Continuous Random Variable: Donglei Du (Ddu@unb - Edu)

Uploaded by

Copyright:

Available Formats

Lecture 7: Continuous Random Variable: Donglei Du (Ddu@unb - Edu)

Uploaded by

Document Information

Original Title

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

Lecture 7: Continuous Random Variable: Donglei Du (Ddu@unb - Edu)

Uploaded by

Copyright:

Available Formats

Lecture 7: Continuous Random Variable

Faculty of Business Administration, University of New Brunswick, NB Canada Fredericton

Donglei Du (UNB) ADM 2623: Business Statistics 1 / 53

A continuous random variable is any random variable whose set of all

Donglei Du (UNB) ADM 2623: Business Statistics 4 / 53

A probability density function (pdf) for any continuous random

Donglei Du (UNB) ADM 2623: Business Statistics 5 / 53

Given a continuous random variable X with its probability density

Geometrically, the probability of B is the area under the curve f (x).

Donglei Du (UNB) ADM 2623: Business Statistics 6 / 53

Consider the continuous random variable X with its probability

For instance, the probability of [1/3, 2/3] is given by

Donglei Du (UNB) ADM 2623: Business Statistics 7 / 53

Donglei Du (UNB) ADM 2623: Business Statistics 8 / 53

When dealing with a continuous random variable, we assume that the

Donglei Du (UNB) ADM 2623: Business Statistics 9 / 53

The standard normal random variable Z has the following probability

Donglei Du (UNB) ADM 2623: Business Statistics 11 / 53

Donglei Du (UNB) ADM 2623: Business Statistics 12 / 53

The pdf is symmetric around its mean x = 0, which is at the same

Donglei Du (UNB) ADM 2623: Business Statistics 13 / 53

A general normal distribution has the following probability density

Donglei Du (UNB) ADM 2623: Business Statistics 14 / 53

Donglei Du (UNB) ADM 2623: Business Statistics 15 / 53

Donglei Du (UNB) ADM 2623: Business Statistics 16 / 53

The pdf is symmetric around its mean x = µ, which is at the same

The 68-95-99.7 (empirical) rule, or the 3-sigma rule: About 68%

Donglei Du (UNB) ADM 2623: Business Statistics 17 / 53

Given X ∼ N (µ, σ 2 ), then

Donglei Du (UNB) ADM 2623: Business Statistics 19 / 53

Donglei Du (UNB) ADM 2623: Business Statistics 21 / 53

Example: Calculate the area between 0 and 1.23.

Donglei Du (UNB) ADM 2623: Business Statistics 23 / 53

Donglei Du (UNB) ADM 2623: Business Statistics 25 / 53

Example: Given the area between 0 and z is 0.3264, find z.

From the table, we find z = 0.94.

Donglei Du (UNB) ADM 2623: Business Statistics 27 / 53

Example: Given the area between less than z is 0.95, find z.

P (Z ≤ z) = 0.95 ⇔ P (0 ≤ Z ≤ z) = 0.95 − 0.5 = 0.45

From the table, we find z = 1.65.

Donglei Du (UNB) ADM 2623: Business Statistics 29 / 53

Donglei Du (UNB) ADM 2623: Business Statistics 31 / 53

Donglei Du (UNB) ADM 2623: Business Statistics 32 / 53

Donglei Du (UNB) ADM 2623: Business Statistics 33 / 53

Donglei Du (UNB) ADM 2623: Business Statistics 34 / 53

Example: Given a normal random variable X ∼ N (50, 82 ), and the

Donglei Du (UNB) ADM 2623: Business Statistics 35 / 53

Donglei Du (UNB) ADM 2623: Business Statistics 36 / 53

Donglei Du (UNB) ADM 2623: Business Statistics 37 / 53

Example: Professor X has determined that the scores in his statistics

Donglei Du (UNB) ADM 2623: Business Statistics 38 / 53

Solution: Let X be the students’ scores. Then X ∼ N (72, 52 ). Let

Donglei Du (UNB) ADM 2623: Business Statistics 39 / 53

Donglei Du (UNB) ADM 2623: Business Statistics 40 / 53

Donglei Du (UNB) ADM 2623: Business Statistics 41 / 53

Example: A manufacturer of aircraft is likely to be very concerned

Donglei Du (UNB) ADM 2623: Business Statistics 42 / 53

Solution: Let X be the the lengths of pilot’s legs. Then

Donglei Du (UNB) ADM 2623: Business Statistics 43 / 53