CHAPTER 4.

DISTRIBUTION THEORY
4.2. PROBABILITY MASS FUNCTION

4.2 Probability Mass Function

Definition 2: Discrete density function of a discrete random variable

If X is a discrete random variable with distinct values x1 , x2 , . . . , xn , . . . , then
the function, denoted by fX (.) and defined by

(
P (X = x) if x = xj , j = 1, 2, . . . , n, . . .
fX (x) = (4.1)
0 if x ̸= xj

is defined to be the discrete density function of X.

• The values of a discrete random variable are often called mass points.
• fX (x) denotes the mass associated with the mass point xj .
• Probability mass function discrete frequency function and probability
function are other terms used in place of discrete density function
• Probability function gives the measure of probability for different values of
X.

4.2.1 Properties of a Probability Mass Function

• Let X be a discrete random variable with probability mass function fX (x).

Then,

1. For any x ∈ ℜ, 0 ≤ fX (x) ≤ 1.

P E be an event and I = {X(ω) : ω ∈ E}. Then P (E) = P (X ∈ I) =
2. Let
x∈I fX (x). P
3. Let R = {X(ω) : ω ∈ Ω}. Then x∈ℜ fX (x) = 1.

4.2.2 Representations of Probability Mass Functions

Example 01
Consider the experiment of tossing a fair coin. Let

(
0 if the outcome is a Tail
X= (4.2)
1 if the outcome is a Head

Find the probability mass function of X. Is X discrete or continuous?

Prepared by Dr. Priyanga D. Talagala (Copyright 2024 P. D. Talagala) 7

4.2. PROBABILITY MASS FUNCTION
CHAPTER 4. DISTRIBUTION THEORY

4.2.2.1 Using a table

8 Prepared by Dr. Priyanga D. Talagala (Copyright 2024 P. D. Talagala)

CHAPTER 4. DISTRIBUTION THEORY
4.2. PROBABILITY MASS FUNCTION

4.2.2.2 Using a function

Prepared by Dr. Priyanga D. Talagala (Copyright 2024 P. D. Talagala) 9

4.3. PROBABILITY DENSITY FUNCTION
CHAPTER 4. DISTRIBUTION THEORY

4.2.2.3 Using a graph

4.3 Probability Density Function

• Let X be a continuous random variable.

• Then, it is not possible to define a pmf fx with properties mentioned in

Section 4.2. Why?

• Instead, we can find a function fx with the some different properties.

• Probability density function (pdf) of a continuous random variable is a

function that describes the relative likelihood for this random variable to
occur at a given point.

10 Prepared by Dr. Priyanga D. Talagala (Copyright 2024 P. D. Talagala)

CHAPTER 4. DISTRIBUTION4.3.
THEORY
PROBABILITY DENSITY FUNCTION

4.3.1 Properties of a Probability Density Function

Let X be a continuous random variable with probability density function fx .

Then,

1. For any x ∈ ℜ, fX (x) ≥ 0.

R E be an event and I = {X(ω) : ω ∈ E}. Then P (E) = P (X ∈ I) =
2. Let
f (x)dx.
I X R
3. Let R = {X(ω) : ω ∈ Ω}. Then ℜ fX (x)dx = 1.

4.3.2 Existence of pdf

• To see the existence of such a function, consider a continuous random

variable X,
• Suppose that we have a very large number of observations, N , of X,
measured to high accuracy (large number of decimal places).
• consider the following grouped frequency table and the histogram con-
structed from those data.
• The height of the bar on a class interval of this histogram is equal to the
relative frequency per unit in that class interval.

Class Class Area of the

Interval boundaries frequency Height of the bar bar
n1 n1
I1 x1 − δx/2, x1 + n1 δx∗N N
δx/2
n2 n2
I2 x2 − δx/2, x2 + n2 δx∗N N
δx/2
.. .. .. .. ..
. . . . .
nk nk
Ik xk − δx/2, xk + nk δx∗N N
δx/2
Total

Prepared by Dr. Priyanga D. Talagala (Copyright 2024 P. D. Talagala) 11

4.3. PROBABILITY DENSITY FUNCTION
CHAPTER 4. DISTRIBUTION THEORY

Figure 4.1: Histograms with different class intervals and a possible model for
the pdf

• Then, for the ith interval,

δx δx
P (xi − ≤ X ≤ xi + ) ≈ Area of the bar
2 2
and therefore

δx δx
Area of the bar P (xi − 2 ≤ X ≤ xi + 2 )
Height of the bar ≈ ≈
δx δx

• Therefore, the height of a bar represents the probability density in that

class interval.
• When δx → 0, it will allow us to approximate the histogram by a smooth
curve as in Figure ?? (d).
• As the area under each histogram is 1, the area under the curve is also 1
• For any point x,

δx δx
P (xi − 2 ≤ X ≤ xi + 2 )
The height of the curve ≈ lim
δx→0 δx
will represent the the probability density at point x.

• Let the above smooth curve be denoted by fX .

• Then, fX has the properties mentioned in Section 4.3.1.

12 Prepared by Dr. Priyanga D. Talagala (Copyright 2024 P. D. Talagala)

CHAPTER 4. DISTRIBUTION4.3.
THEORY
PROBABILITY DENSITY FUNCTION

• The function is called the probability density function of X.

• NOTE Here fX (x) represents Probability density at point x. Not

Probability at point x.

4.3.3 Calculation of Probability using pdf

• Let c, d ∈ ℜ such that c ≤ d. Then,

Z d
P (c ≤ X ≤ d) = fX (x)dx
c

Rd
Figure 4.2: P (c ≤ x ≤ d = c
fX (x)dx

• NOTE: if X is a continuous random variable with the p.d.f fX , then for

any k ∈ ℜ,

Z k
P (X = k) = P (k ≤ X ≤ k) = fX (x)dx = 0
k

• Therefore, for a continuous random variable X,

Z d
P (c < X < d) = P (c ≤ X < d) = P (c < X ≤ d) = P (c ≤ X ≤ d) = fX (x)dx
c

Prepared by Dr. Priyanga D. Talagala (Copyright 2024 P. D. Talagala) 13

4.4. CUMULATIVE DISTRIBUTION
CHAPTER
FUNCTION
4. DISTRIBUTION THEORY

4.4 Cumulative Distribution Function

• There are many problems in which it is of interest to know the probability
that the values of a random variable is less than or equal to some real
number x.

Definition 3: Cumulative distribution function

The cumulative distibution function or cdf of a random variable X, denoted by
FX (x), is defined by

FX (x) = P (X ≤ x), for all x

4.4.1 Relationship between cdf and pmf

• Therefore, if X is a discrete random variable, the cdf is given by,

X
FX (x) = fX (t), − ∞ < x < ∞
t≤x

where fX (t) is the value of the pmf of X at t.

Example 02
An owner of a software engineering company is interested in knowing how many
years his employees stay with his company. Let X be the number of years an
employee will stay with the company. Over the years, he has established the
following probability distribution:

14 Prepared by Dr. Priyanga D. Talagala (Copyright 2024 P. D. Talagala)

CHAPTER 4. DISTRIBUTION
4.4. CUMULATIVE
THEORY DISTRIBUTION FUNCTION

x 1 2 3 4 5 6 7
fX (x) = P (X = x) 0.1 0.05 0.1 ? 0.3 0.2 0.1

1. Find fX (4)

2. Find P (X < 4)

Prepared by Dr. Priyanga D. Talagala (Copyright 2024 P. D. Talagala) 15

4.4. CUMULATIVE DISTRIBUTION
CHAPTER
FUNCTION
4. DISTRIBUTION THEORY

3. Find P (X ≤ 4)

4. Draw the probability mass function of X

16 Prepared by Dr. Priyanga D. Talagala (Copyright 2024 P. D. Talagala)

CHAPTER 4. DISTRIBUTION
4.4. CUMULATIVE
THEORY DISTRIBUTION FUNCTION

5. Draw the cumulative distribution function of X

Prepared by Dr. Priyanga D. Talagala (Copyright 2024 P. D. Talagala) 17

4.4. CUMULATIVE DISTRIBUTION
CHAPTER
FUNCTION
4. DISTRIBUTION THEORY

4.4.1.1 Properties of a cumulative distribution function of a Discrete

random variable

• It is a step function
• It is a non decreasing function that increases from 0 to 1

4.4.2 Relationship between cdf and pdf

• If X is a continuous random variable, the cdf is given by,

Z x
FX (x) = fX (t)dt − ∞ < x < ∞
−∞

where fX (t) is the value of the pdf of X at t. (Here t is a dummy integration

variable).

• Conversely,

dFX (x)
fX (x) =
dx
Example 03

1
 25 x
 0≤x<5
2 1
fX (x) = 5 − 25 x 5 ≤ x ≤ 10 (4.3)

0 otherwise


1. Find the CDF of X

2. Find P (X ≤ 8)
3. Find P (3 ≤ X ≤ 8)

18 Prepared by Dr. Priyanga D. Talagala (Copyright 2024 P. D. Talagala)

CHAPTER 4. DISTRIBUTION
4.4. CUMULATIVE
THEORY DISTRIBUTION FUNCTION

1
.

Prepared by Dr. Priyanga D. Talagala (Copyright 2024 P. D. Talagala) 19