Random Variables

Mahbub Latif, PhD

January 2025
Plan
Introduction
Discrete random variables
Continuous random variables
Expectation of a random variable
Variance of a random variable
Jointly distributed random variables

2
Random variables

Random variables are one of the fundamental building blocks of probability

theory and statistical inference
A random variable is formed by assigning a numerical value to each
outcome in the sample space of a particular experiment
A random variable can be thought of as being generated from a function
that maps each outcome in a particular sample space onto the real number
line R

3
A random variable is obtained by assigning a numerical value to each
outcome of a particular experiment.
4
Example 1 (Machine breakdowns)

repair cost
outcome probability (in USD)
electrical 0.2 200
mechanical 0.5 350
misuse 0.3 50

Repair cost is a random variable as it is a numeric value and it corresponds to

each element of the sample space

5
Example 4 (power plant operation)

X = no. of power plants working and its possible values are 0, 1, 2, and 3

6
7
Types of random variables

A random variable is either discrete or continuous

Number of power plants generating electricity ⟶ a discrete random
variable
Lifetime of a laptop battery ⟶ a continuous random variable

8
Types of random variables

Random variables are generally denoted by uppercase letters, such as X, Y ,

Z, etc.

Lowercase letters (e.g. x, y, etc.) are used to denote values taken by the
random variable
E.g. X denote the number of power plants generating electricity and its
values are denoted by lowercase letters x = 0, x = 1, etc.

9
Homework 2B

2.2.1 Consider a random variable measuring the following quantities. In each

case state with reasons whether you think it more appropriate to define the
random variable as discrete or as continuous.
A person’s height
A student’s course grade (CGPA, point-grade, grade out of 100)
The thickness of a metal plate
A person’s age

10
Discrete random variables

11
Probability mass function

Probability mass function (pmf) is defined for a discrete random variable and
it assigns probability values to all possible values of the random variable
The probability mass function of a discrete random variable X is a set of
probability values p assigned to values of random variable x
i

P (X = x) = px

Probability values must satisfy

(i) 0 ≤ px ≤ 1 and (ii) ∑ px = 1

x

12
Probability distribution of repair costs of machine breakdown

outcome repair cost,

xi probability,pi

electrical 200 0.2

mechanical 350 0.5
misuse 50 0.3

13
Probability distribution of the number of plants generating electricity

xi 0 1 2 3
pi 0.07 0.23 0.57 0.13

14
Cumulative distribution function

The cumulative distribution function of a random variable X is defined as

F (a) = P (X ≤ a)

= ∑ P (X = y)

y: y≤a

Like the probability mass function, the cumulative distribution function

summarizes the probabilistic properties of a random variable.
Knowledge of either the probability mass function or the cumulative
distribution function allows the other functions to be calculated.

15
Probability distribution of repair costs of machine breakdown

outcome repair cost,

xi probability,pi

electrical 200 0.2

mechanical 350 0.5
misuse 50 0.3

−∞ < x < 50 ⇒ F (x) = P (cost ⩽ x) = 0

50 ≤ x < 200 ⇒ F (x) = P (cost ⩽ x) = .30

200 ≤ x < 350 ⇒ F (x) = P (cost ⩽ x) = .50

350 ≤ x < ∞ ⇒ F (x) = P (cost ⩽ x) = 1.0

16
Cumulative distribution function

For a discrete random variable, F (x) is an increasing step function with

steps at the values taken by the random variable

17
Cumulative distribution function

Probability mass function (pmf) can be obtained from cumulative

distribution function (cdf)
−
P (X = x) = F (x) − F (x )

F (x is the limiting value from below of the cumulative distribution

−
)

function
If there is no step in the cumulative distribution function at a point x
−
F (x) = F (x ) and P (X = x) = 0

18
 −
P (X = 40) = F (40) − F (40 ) = 0
−
P (X = 50) = F (50) − F (50 ) = .30 − 0 = .30

−
P (X = 65) = F (65) − F (65 ) = .30 − .30 = 0

19
Homework 2A

2.1.1 An office has four copying machines, and the random variable X
measures how many of them are in use at a particular moment in time.
Find P (X = 4) for the given probabilities
P (X = 0) = 0.08, P (X = 1) = 0.11,

P (X = 2) = 0.27, P (X = 3) = 0.33

Draw a line graph of the probability mass function

Construct and plot the cumulative distribution function

20
Homework 2A

2.1.3 Suppose that two fair dice are rolled and that the two numbers recorded
are multiplied to obtain a final score.
Construct and plot the probability mass function and the cumulative
distribution function of the final score

21
Continuous Random Variables

22
Example 14 (Metal Cylinder Production)

A company manufactures metal cylinders, which are designed to have a

diameter of 50 mm, but the company discovers that the cylinders can have a
diameter anywhere between 49.5 and 50.5 mm
Suppose that the random variable X is the diameter of a randomly chosen
cylinder manufactured by the company
Since this random variable can take any value between 49.5 and 50.5, it is a
continuous random variable

23
Example 15 (Battery failure times)

Suppose that a random variable X is the time to failure of a newly charged

battery
Failure can be defined to be the moment at which the battery can no longer
supply enough energy to operate a certain appliance
This random variable is continuous since it can hypothetically take any
positive value.
Its state space can be thought of as the interval 0 to ∞

24
Example 17 (Milk contents)

A machine-filled milk container is labeled as containing 2 liters.

However, the actual amount of milk deposited into the container by the
filling machine varies between 1.95 and 2.20 liters.
If the random variable X measures the amount of milk in a randomly
chosen container, it is a continuous random variable taking any value in the
interval [1.95, 2.20]

25
Probability density function

The main distinction between discrete and continuous random variables lies
in how their probabilistic properties are defined.
The probabilistic properties of discrete random variables are defined
through a probability mass function
The probabilistic properties of a continuous random variable are defined
through a probability density function

26
Probability density function

The probabilistic properties of a continuous random variable are defined

through a function f (x)

A function f (x) is said to be a density function if it satisfies the following

two properties:
f (x) > 0 for all values of x
∞
∫ f (x)dx = 1, −∞ < x < ∞
−∞

27
Probability density function

The probability that a continuous random variable lies between two values a
and b is obtained by integrating the probability density function between
these two values

P (a ≤ X ≤ b) = P (a < X < b)

= ∫ f (x) dx
a

28
Probability density function

The probability that a continuous random variable takes a specific value is

zero, i.e P (X = a) = 0
a

P (X = a) = ∫ f (x) dx = 0
a

29
Example 15 (Battery failure times)

Suppose battery failure times X (measured in hours) has a probability

density function
2
3
x ≥ 0
(1+x)
f (x) = {
0 x < 0

30
Example 15 (Battery failure times)

It can be shown that f (x) > 0 for all x ≥ 0 and

∞ ∞ ∞
2 −1 ∣
∫ f (x) dx = ∫ dx = ∣ = 1
3 2
0 0
(1 + x) (1 + x) ∣
0

f (x) is a valid probability density function

What is the probability that the battery fails within first five hours?

31
Example 15 (Battery failure times)

What is the probability that the battery fails within first five hours?
5 5
2 −1 ∣ −1 35
P (X ≤ 5) = ∫ dx = ∣ = + 1 =
3 2 2
0
(1 + x) (1 + x) ∣ 6 36
0

What is the probability that a battery lasts longer than five hours?

32
Example 17 (Milk container contents)

Suppose that the probability density function of the amount of milk

deposited in a milk container is
−x
f (x) = 40.976 − 16x − 30e , 1.95 ≤ x ≤ 2.20

33
Example 17 (Milk container contents)

It can be shown that

−1.95
f (1.95) = 40.976 − (16)(1.95) − (30)(e ) = 5.508 > 0

f (2.20) = 40.976 − (16)(2.20) − (30)(e

−2.20
) = 2.452 > 0 and
2.20 2.20
∣
−x 2 −x
∫ (40.976 − 16x − 30e ) dx = (40.976x − 16(x /2) + 30e )∣ = 1
1.95
∣
1.95

34
The probability that the actual amount of milk is less than 2.0 liter
2.0 2.0

−x
∫ f (x) dx = ∫ (40.976 − 16x − 30e ) dx
1.95 1.95

2.0
∣
2 −x
= (40.976x − 16(x /2) + 30e )∣
∣
1.95

2 −2.0
= [(40.976)(2.0) − 16(2.0 /2) + 30e ]

2 −1.95
− [(40.976)(1.95) − 16(1.95 /2) + 30e ]

= 54.012 − 53.751 = 0.261

About 26% of the milk containers are underweight.

35
Cumulative Distribution Function

The cumulative distribution function of a continuous random variable X is

defined in exactly the same way as for a discrete random variable
a

F (a) = P (X ≤ a) = ∫ f (x) dx
−∞

f (x) → probability density function of X

36
Cumulative Distribution Function

For a continuous random variable, the probability density function can also
be obtained from cumulative distribution function
dF (x)
f (x) =
dx

Probability that a continuous random variable X lies between a and b can

be obtained using cumulative distribution function
P (a ≤ X ≤ b) = P (X ≤ b) − P (X ≤ a) = F (b) − F (a)

37
Example 15 (Battery failure times)

Probability density function

2
f (x) = x ≥ 0
3
(1 + x)

38
Cumulative distribution function
x

F (x) = ∫ f (y) dy
0
x
2
= ∫ dy
3
0
(1 + y)

1
= 1 −
2
(1 + x)

39
Example 15 (Battery failure timesY)

Probability density function Cumulative distribution function

2 1
f (x) = x ≥ 0 F (x) = 1 −
3 2
(1 + x) (1 + x)

40
Example 15 (Battery failure timesY)

Find the probability that a battery lasts between one and two hours
2

P (1 ≤ X ≤ 2) = ∫ f (x) dx
1

2 2
2 −1 ∣ 1 1
= ∫ dx = ∣ = − = 0.21
(1 + x)
3
(1 + x)
2 ∣ 4 25
1 1

P (1 ≤ X ≤ 2) = F (2) − F (1)

1 1 1 1
= [1 − ] − [1 − ] = − = 0.21
2 2
(1 + 2) (1 + 1) 4 25

41
Homework 2B

2.2.2 A random variable X takes values between 4 and 6 with a probability

density function
1
f (x) = , 4 ≤ x ≤ 6.
x ln(1.5)

Check that the total area under the probability density function is equal to 1.
What is P (4.5 ≤ X ≤ 5.5) ?

Construct the cumulative distribution function.

42
Homework 2B

2.2.4 A random variable X takes values between 0 and 4 with a cumulative

distribution function
2
x
F (x) = f or 0 ≤ x ≤ 4
16

What is P (X ≤ 2) ?

What is P (1 ≤ X ≤ 3) ?

What is the probability density function.

43
Homework 2B

2.2.6 A car panel is spray-painted by a machine, and the technicians are

particularly interested in the thickness of the resulting paint layer.
Suppose that the random variable X measures the thickness of the paint in
millimeters at a randomly chosen point on a randomly chosen car panel, and
that X takes values between 0.125 and 0.5 mm with a probability density
function of
2
f (x) = A[0.5 − (x − 0.5) ], 0.125 ≤ x ≤ 0.5

Find the value of A and construct the cumulative distribution function.

What is the probability that the paint thickness at a particular point is less
than 0.2 mm? 44
The Expectation of a Random Variable

45
The Expectation of a Random Variable

The probability mass function or the probability density function provides

complete information about the probabilistic properties of a random
variable
Summary measures of a random variable would be useful, two commonly
used summary measures are expectation and variance of a random variable
Expectation represents the "average" value of the random variable, where as
variance measures average distance of a variable from its mean value
Expectation of a random variable X is denoted by E(X)

46
Expectation of a discrete random variable

The expected value or expectation of a discrete random variable X with a

probability mass function P (X = x ) = p is defined as
i i

μ = E(X) = ∑ P (X = xi ) xi = ∑ pi xi

i i

E(X) provides a summary measure of the average value taken by the

random variable and is also known as the mean of the random variable

47
Example 1 (Machine breakdowns)

repair cost
outcome probability (in USD)
electrical 0.2 200
mechanical 0.5 350
misuse 0.3 50

Expected repair cost

E(X) = ∑ pi xi = (.2)(200) + (.5)(350) + (.3)(50) = 230

48
Example 4 (power plant operation)

xi 0 1 2 3
pi 0.07 0.23 0.57 0.13

Expected number of plants generating electricity

E(X) = ∑ pi xi = (.07)(0) + (.23)(1) + (.57)(2) + (.13)(3) = 1.76

49
Expectation of Continuous Random Variables

The expected value or expectation of a continuous random variable X with

a probability density function f (x) is defined as

E(X) = ∫ xf (x) dx
x

50
Example 15 (Battery Failure Times)

∞ ∞
2 2
E(X) = ∫ x dx E(X) = ∫ x dx
3 3
0
(1 + x) 0
(1 + x)

∞ ∞
2 2 2(y − 1)
= ∫ [ − ]dx = ∫ [ ]dy
2 3 3
0
(1 + x) (1 + x) 1
y

∞ ∞

−2 1 2 2
= [ + ] = [ − ]
2 2 3
(1 + x) (1 + x) y y
0 1

= 1 = 1

51
Properties of expectation

For two constants a and b

E(a) = a

E(aX) = aE(X)

E(a + bX) = a + bE(X)

52
Homework 2C

2.3.1 An office has four copying machines, and the random variable X
measures how many of them are in use at a particular moment in time.
Suppose that
P (X = 0) = 0.08

P (X = 1) = 0.11

P (X = 2) = 0.27

P (X = 3) = 0.33.

What is the expected number of copying machines in use at a particular

moment in time?

53
Homework 2C

2.3.11 Consider again the random variable with a cumulative distribution

function of
2
F (x) = x /16; 0 ≤ x ≤ 4

What is the expected value of this random variable?

54
Homework 2C

2.3.12 Consider again the car panel painting machine discussed in Problem
2.2.6, where
2
f (x) = A[0.5 − (x − 0.5) ], 0.125 ≤ x ≤ 0.5

What is the expected paint thickness?

55
The Variance of a Random Variable

56
The Variance of a Random Variable

Variance of a random variable measures variability or spread in the values

taken by the random variable
The variance of a random variable X is denoted by σ , and is defined as
2

2
2
σ = V ar(X) = E[X − E(X)]

2 2
= E[X − 2XE(X) + E(X) ]

2 2 2
= E(X ) − 2E(X) + E(X )

2 2
= E(X ) − E(X)

2 2
= E(X ) − μ

57
Standard deviation

The positive square root of the variance is known as the standard deviation,
which is denoted by σ
2
σ = +√ σ = +√V ar(X)

58
Example 1 (Machine breakdowns)

x, repair cost p=P(X=x) x2 px px2

50 0.3 2,500 15 750
200 0.2 40,000 40 8,000
350 0.5 122,500 175 61,250
Total 1.0 165,000 230 70,000

59
Expected value

E(X) = ∑ xi pi = 230

Variance
2 2
V ar(X) = E(X ) − E(X)

2 2
= ∑ x pi − E(X)
i

2
= 70, 000 − 230 = 17, 100

Standard deviation

σ = √V ar(X) = √17, 100 = 130.767

60
Example 14 (Metal cylinder diameter)

The probability density function of metal cylinder diameter

2
f (x) = 1.5 − 6(x − 50.0) , 49.5 ≤ x ≤ 50.5

Expected value
50.5 .5

2 2
E(X) = ∫ x[1.5 − 6(x − 50.0) ]dx = ∫ (y + 50)(1.5 − 6y )dy
49.5 −.5

.5

3 2
= ∫ (1.5y − 6y + 75 − 300y )dy
−.5

.5
2 4 3
= [(1.5y /2) − (6y /4) + 75y − (300y /3)] = 50
−.5
61
Example 14 (Metal cylinder diameter)

The probability density function of metal cylinder diameter

2
f (x) = 1.5 − 6(x − 50.0) , 49.5 ≤ x ≤ 50.5

Variance
2 2 2
V ar(X) = E(X ) − E(X) ] = 2500.05 − 50 = 0.05

62
 .5

2 2 2
E(X ) = ∫ (y + 100y + 2500)(1.5 − 6y )dy
−.5

.5

2 4 3 2
= ∫ [1.5y + 150y + 3750 − 6y − 600y − 15000y ]dy
−.5

0.5
3 2 5 4 3
= [.5y + 75y + 3750y − (6y /5) − (600y /4) − 5000y ]
−.05

= 1259.4 − (−1240.65) = 2500.05

63
Properties of variance

For two constants a and b

V (a) = 0

2
V (aX) = a V (X)

2
V (a + bX) = b V (X)

64
Quantiles

65
Medians of Random Variables

The median is another summary measure of the distribution of a random

variable that provides information about the "middle" value of the random
variable
The median of a continuous random variable X is x (say), which satisfies
m

F (xm ) = 0.5

66
Medians of Random Variables

The cumulative distribution function of battery lifetime is

1
F (x) = 1 −
2
(1 + x)

The median battery lifetime

1
F (x) = 0.5 ⇒ 1 − = 0.5
2
(1 + x)

2
⇒ (1 + x) = 2

⇒ x = √2 − 1 = 0.41

67
Symmetric distribution

A continuous random variable X is said to be symmetric about a point μ if

both the median and the expectation of the random variable are equal to μ

68
Quantiles

Quantiles of random variables are additional summary measures that can

provide information about the spread or variability of the distribution of the
random variable
The pth quantile (0 < p < 1) of a random variable X with a cumulative
distribution function F (x) is defined to be the value x for which
p

F (xp ) = P (X ≤ xp ) = p

It is also referred to as the p × 100th percentile of the random variable

There is a probability of p that the random variable takes a value less than
the pth quantile xp

69
Quartiles

The .25 quantile (x .25 ) is the first quartile Q 1

The .75 quantile (x .75 ) is the third quartile Q 3

The median (x .5 ) is the second quartile Q 2

70
Inter-quartile range

The inter-quartile range (IQR) is defined as the difference between the

third and first quartile
I QR = x.75 − x.25 = Q3 − Q1

IQR is a measure of spread

71
Quartiles

72
Example 15 (Battery Failure Times)

The cumulative distribution function of battery lifetime

1
F (x) = 1 − , x ≥ 0
2
(1 + x)

73
Example 15 (Battery Failure Times)

Third and first quartiles

1
F (x.75 ) = .75 ⇒ 1 − = .75
2
(1 + x.75 )

1
⇒ = 0.25 ⇒ x.75 = 1 hr
2
(1 + x.75 )

1
F (x.25 ) = .25 ⇒ 1 − = .25
2
(1 + x.25 )

1
⇒ = 0.75 ⇒ x.25 = .154 hr
2
(1 + x.25 )

74
Example 15 (Battery Failure Times)

Interquartile range (IQR)

I QR = x.75 − x.25 = 1 − 0.154 = 0.845

Half of the batteries will fail between 0.154 hour and 1 hour (i.e. between
about 9 minutes to 60 minutes)

75
Homework 2D

2.4.1 Suppose that the random variable X takes the values −2, 1, 4, and 6 with
probability values 1/3, 1/6, 1/3, and 1/6, respectively.
Find the expectation and variance of X.

76
Homework 2D

2.4.5 Consider again the random variable described in Problems 2.2.2 and
2.3.10 with a probability density function of
1
f (x) = , 4 ≤ x ≤ 6.
x ln(1.5)

What is the variance of this random variable?

What is the standard deviation of this random variable?
Find the upper and lower quartiles of this random variable.
What is the interquartile range?
77
Homework 2D

2.4.6 Find variance, standard deviation, and interquartile range of the variable
X , where

2
x
F (x) = f or 0 ≤ x ≤ 4
6

78
Jointly Distributed Random Variables

79
Jointly Distributed Random Variables

For two discrete random variables X and Y , the joint probability mass
function is defined by
P (X = xi , Y = yj ) = pij

0 < pij < 1 for all i and j

∑ ∑ pij = 1
i j

80
Jointly Distributed Random Variables

For two continuous random variables X and Y , the joint probability density
function is defined by f (x, y) that satisfies

f (x, y) > 0 for all x and y

∫ ∫ f (x, y) dy dx = 1
x y

81
Jointly Distributed Random Variables

The probability that X lies between a and b, and Y lies between c and d is
defined as
b d

P (a ≤ X ≤ b, c ≤ Y ≤ d) = ∫ ∫ f (x, y) dy dx
a c

The joint cumulative distribution function is defined as

x y

F (x, y) = P (X ≤ x, Y ≤ y) = ∫ ∫ f (u, v) dx dy
−∞ −∞

82
Example 19 (Air Conditioner Maintenance)

A company that services air conditioner units is interested in how long a

technician takes on a visit to a particular location, which depends on the
number of air conditioner units at the location that need to be serviced
X ∈ {1, 2, 3, 4} → the service time in hours taken at a particular
location
Y ∈ {1, 2, 3} → the number of air conditioner units at the location

The random variables X and Y are jointly distributed

83
Example 19 (Air Conditioner Maintenance)

P (X = 1, Y = 1) = p11 = 0.12

It is a valid joint mass function because all probabilities are non-negative

and sum of all probabilities is 1, i.e. ∑ ∑ p = 1
i j ij
84
Example 19 (Air Conditioner Maintenance)

P (X ≤ 1, Y ≤ 2) = p11 + p12

= .12 + .08 = .20

85
Marginal probability distributions

The marginal distribution is the individual probability distribution of the

random variable, which can be obtained from a joint distribution
For a discrete random variable

P (X = xi ) = ∑ P (X = xi , Y = yj )

E.g. P (X = 1) = ∑ P (X = 1, Y = yj )

j=1

= p11 + p12 + p13 + p14 = 0.21

86
87
Marginal probability distributions

Marginal distribution of a continuous variable

∞

f1 (x) = ∫ f (x, y) dy
−∞

f2 (y) = ∫ f (x, y) dx
−∞

88
Marginal probability distributions

Marginal distribution of a discrete variable

P (X = x) = ∑ P (X = x, Y = y)
y

P (Y = y) = ∑ P (X = x, Y = y)
x

89
Conditional probability distribution

The conditional distribution of a random variable X conditional on a

random variable Y taking a particular value is defined as
P (X = x, Y = y)
Px | y
= P (X = x | Y = y) = (discrete)
P (Y = y)

f (x, y)
f2 | 2
(x|y) = (continuous)
f (y)

90
Conditional probability distribution

For example
P (X = 2, Y = 1) .08
P (Y = 1 | X = 2) = = = .33
P (X = 2) .24

91
Independent random variables

Two random variables X and Y are defined to be independent if their joint

probability mass function or joint probability density function is the product
of their two marginal distributions
P (X = xi , Y = Yj ) = P (X = xi ) P (Y = yj ) [discrete]

f (x, y) = f (x) f (y) [continuous]

92
Independent random variables

Are X and Y independent for f (x, y) = 6xy 2

, (0 ≤ x ≤ 1, 0 ≤ y ≤ 1)?

1 1
1
2 3
f (x) = ∫ f (x, y)dy = ∫ 6xy dy = [6xy /3] = 2x
0
0 0

1 1
1
2 2 2 2
f (y) = ∫ f (x, y)dx = ∫ 6xy dx = [6x y /2] = 3y
0
0 0

Since f (x) f (y) = (2x)(3y 2

) = f (x, y)

So X and Y are independent

93
Homework 2E

2.5.3 Suppose that two continuous random variables X and Y have a joint
probability density function
f (x, y) = A(x − 3)y, −2 ≤ x ≤ 3, 4 ≤ y ≤ 6

What is the value of A?

What is P (0 ≤ X ≤ 1, 4 ≤ Y ≤ 5) ?

Construct the marginal probability density functions of X and Y .

Are the random variables X and Y independent?

94
 3 6

∫ ∫ A(x − 3)y dy dx = 1
−2 4

3 6
2 ∣
A∫ (x − 3)[(y /2)∣ ]dx = 1
∣
−2 4

10A ∫ (x − 3) dx = 1
−2

3
2∣
10A(x /2 − 3x) = 1
∣
−2

10A[(9/2) − 9 − 2 − 6] = 1

10A(−25/2) = 1

A = −(1/125)

95