UNIT-I THE RANDOM VARIABLE

UNIT-I
THE RANDOM VARIABLE

The Random Variable: Concept of random variable, Condition for a function

to be a random variable, Classification of a random variable, Cumulative
distribution function and properties, Probability density function and properties,
Different distributions-Binomial, Poisson, Uniform, Exponential, Rayleigh,
Gaussian.
Operations on One Random Variable: Expectation, Moments about the
origin, Central moments, Variance, Skew, Skewness, Characteristic function
and properties, Moment generating function and properties, Transformations of
a random variable.

PROBABILITY
 Probability theory is used in all those situations where there is
randomness about the occurrence of an event.
(Or)
Measure of the likeliness that an event will occur.
 The meaning of randomness is existence of certain amount of unsureness
about an event.
 This section deals with the chances of occurring any particular
phenomena. i.e. Electron emission, Telephone calls, Radar detection,
quality control, system failure, games of chance, birth and death rates,
Random walks, probability of detection, Probability of false alarm, BER
calculation, optimal coding (Huffman) and many more.
Experiment:-
A random experiment is an action or process that leads to one of several
possible outcomes

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 1

 UNIT-I THE RANDOM VARIABLE

Sample Space:
The sample space is the collection of all possible outcomes of a random
experiment. The elements of are called sample points.
 A sample space may be finite, countably infinite or uncountable.
 A finite or countably infinite sample space is called a discrete
sample space.
 An uncountable sample space is called a continuous sample space
Types of Sample Space:
Finite/Discrete Sample Space:
Consider the experiment of tossing a coin twice. The sample space can be
S = {HH, HT, T H , TT} the above sample space has a finite number of
sample points. It is called a finite sample space.
Countably Infinite Sample Space:
 Consider that a light bulb is manufactured. It is then tested for its life
length by inserting it into a socket and the time elapsed (in hours) until it
burns out is recorded.
 Let the measuring instrument is capable of recording time to two decimal
places, for example 8.32 hours.
 Now, the sample space becomes count ably infinite i.e.
S = {0.0, 0.01, 0.02}
The above sample space is called a countable infinite sample space.
Uncountable / Infinite Sample Space/Continuous Sample Space
If the sample space consists of unaccountably infinite number of
elements then it is called uncountable/ Infinite Sample Space.
Event
An event is simply a set of possible outcomes. To be more specific, an
event is a subset A of the sample space S.
Example 1: tossing a fair coin
Example 2: Throwing a fair die

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 2

 UNIT-I THE RANDOM VARIABLE

Types of Events:
Exhaustive Events:
 A set of events is said to be exhaustive, if it includes all the possible
events.
Ex. In tossing a coin, the outcome can be either Head or Tail and there is
no other possible outcome. So, the set of events { H , T } is exhaustive.

Mutually Exclusive Events:

 Two events, A and B are said to be mutually exclusive if they cannot
occur together.
If two events are mutually exclusive then the probability of either
occurring is

Ex. In tossing a die, both head and tail cannot happen at the same time.

Equally Likely Events:

 Each outcome of the random experiment has an equal chance of
occurring.
Ex. In tossing a coin, the coming of the head or the tail is equally likely.

Independent Events:
 Two events are said to be independent, if happening or failure of one
does not affect the happening or failure of the other. Otherwise, the
events are said to be dependent.
 If two events, A and B are independent then the joint probability is

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 3

 UNIT-I THE RANDOM VARIABLE

Axioms of Probability
For any event A, we assign a number P(A), called the probability of the event
A. This number satisfies the following conditions that act the axioms of
probability.
(i) P( A) ≥ 0 (Probabili ty is a nonnegativ e number)
(ii) P(s) = 1 (Probabili ty of the whole set is unity)

Note that (iii) states that if A and B are mutually exclusive (M.E.) events, the
probability of their union is the sum of their probabilities
Relative Frequency
 Random experiment with sample space S. we shall assign non-negative
number called probability to each event in the sample space.
 Let A be a particular event in S. then “the probability of event A” is
denoted by P(A).
 Suppose that the random experiment is repeated n times, if the event A
occurs nA times, then the probability of event A is defined as "Relative
frequency"
• Relative Frequency Definition:
The probability of an event A is defined as

Joint probability
 Joint probability is defined as the probability of joint (or) simultaneous
occurrence of two (or) more events.
 Let A and B taking place, and is denoted by P (AB) or P(A∩B)
P (AB) = P(A∩B = P(A) + P(B) - P(AUB)

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 4

 UNIT-I THE RANDOM VARIABLE

Conditions for a function to be a Random Variable

 Every point in sample space must be correspond only one value of the
random variable.
 The set {𝑋 ≤ 𝑥 } shall be an event for any real number 𝑥.The probability
of this event is equal to the sum of the probabilities of all the elementary
events corresponding to{𝑋 ≤ 𝑥 }. This is denoted as P{𝑋 ≤ 𝑥 }
 The probability of events {𝑋 ≤ −∞} 𝑎𝑛𝑑 {𝑋 ≤ ∞} are zero.

PROBABILITY DISTRIBUTION FUNCTION

 Probability distribution function 𝐹𝑋 (𝑥 ) gives information about the
probabilistic behaviour of a random variable.
 It gives information about probability of event {𝑋 ≤ 𝑥 }
𝐹𝑋 (𝑥 ) = P{𝑋 ≤ 𝑥 }
 Probability distribution function is also called as Cumulative distribution
function (CDF).

Properties of Probability Distribution Function

1. The Probability distribution function is zero at
𝑥 = −∞ 𝑖. 𝑒; 𝐹𝑋 (−∞) = 0
Proof:
It is known that
𝐹𝑋 (𝑥 ) = P{𝑋 ≤ 𝑥 }
𝐹𝑋 (−∞) = P{𝑋 ≤ −∞}
𝐹𝑋 (−∞) = 0
2. The Probability distribution function is unity at 𝑥 = ∞
𝑖. 𝑒; 𝐹𝑋 (∞) = 1
Proof:
It is known that

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 5

 UNIT-I THE RANDOM VARIABLE

𝐹𝑋 (𝑥 ) = P{𝑋 ≤ 𝑥 }
𝐹𝑋 (∞) = P{𝑋 ≤ ∞}
𝐹𝑋 (∞) = P{𝑋 ≤ −∞} + ⋯ + P{𝑋 ≤ 0} + P{𝑋 ≤ 1} + ⋯ P{𝑋 ≤ ∞}
𝐹𝑋 (∞) = 1

3. The Probability distribution function is always define between 0 and 1.

i.e; 0 ≤ 𝐹𝑋 (𝑥 ) ≤ 1

4. The Probability distribution function is non decreasing function

i.e; 𝐹𝑋 (𝑥2 ) ≥ 𝐹𝑋 (𝑥1 ) 𝑤ℎ𝑒𝑛 𝑥2 > 𝑥1

5. The Probability of an event 𝑥1 < 𝑋 ≤ 𝑥2 can be obtained from the

knowledge of Probability distribution function
𝑃{𝑥1 < 𝑋 ≤ 𝑥2 } = 𝐹𝑋 (𝑥2 ) − 𝐹𝑋 (𝑥1)
Proof
Given that 𝑥2 > 𝑥1
{𝑋 ≤ 𝑥2 } = {𝑋 ≤ 𝑥1 } + {𝑥1 ≤ 𝑋 ≤ 𝑥2 }
Adding probabilities to the above events
𝑃 {𝑋 ≤ 𝑥2 } = 𝑃 {𝑋 ≤ 𝑥1} + 𝑃 {𝑥1 ≤ 𝑋 ≤ 𝑥2 }
𝑃{𝑥1 ≤ 𝑋 ≤ 𝑥2 } = 𝑃{𝑋 ≤ 𝑥2 } − 𝑃 {𝑋 ≤ 𝑥1 }
𝑃{𝑥1 ≤ 𝑋 ≤ 𝑥2 } = 𝐹𝑋 (𝑥2 ) − 𝐹𝑋 (𝑥1 )

Probability Density Function:

 Probability density function will give the information about the probability
of individual outcomes P{𝑋 = 𝑥 }
 Let F(x) be the distribution function for a continuous random variable X.
The probability density function (PDF) for X is given by
𝑑𝐹𝑋 (𝑥 )
𝑓𝑋 (𝑥 ) =
𝑑𝑥
CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 6
 UNIT-I THE RANDOM VARIABLE

Properties of Probability density function

1. Density function is a non-negative quantity
𝑓𝑋 (𝑥 ) ≥ 𝑜
Proof
𝑑𝐹𝑋 (𝑥)
From the definition 𝑓𝑋 (𝑥 ) =
𝑑𝑥

As the distribution function is anon non-decreasing function slope is always

positive. Hence the probability density function is a non-negative quantity.

2. The area under the probability density function is unity.

∞
∫ 𝑓𝑋 (𝑥 )𝑑𝑥 = 1
−∞

Proof:
∞ ∞
𝑑
∫ 𝑓𝑋 (𝑥 )𝑑𝑥 = ∫ 𝐹𝑋 (𝑥 )
−∞ −∞ 𝑑𝑥

[𝐹𝑋 (𝑥 )]∞
−∞ = 𝐹𝑋 (∞ ) − 𝐹𝑋 (−∞ )

=1−0 =1
3. The probability distribution function can be obtained from the
knowledge of density function. It means distribution function is the area
under the density function.
𝑥
𝐹𝑋 (𝑥 ) = ∫ 𝑓𝑋 (𝑥 )𝑑𝑥
−∞

Proof
𝑑
𝑓𝑋 (𝑥 ) = 𝐹 (𝑥 )
𝑑𝑥 𝑋
Integrating on both sides
𝑥 𝑥
𝑑
∫ 𝑓𝑋 (𝑥 )𝑑𝑥 = ∫ 𝐹𝑋 (𝑥 )
−∞ −∞ 𝑑𝑥
𝑥
= [𝐹𝑋 (𝑥 )]−∞

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 7

 UNIT-I THE RANDOM VARIABLE

= 𝐹𝑋 (𝑥 ) − 𝐹𝑋 (−∞) = 𝐹𝑋 (𝑥 )
𝑥
𝐹𝑋 (𝑥 ) = ∫−∞ 𝑓𝑋 (𝑥 )𝑑𝑥
4. The Probability of an event {𝑥1 < 𝑋 ≤ 𝑥2 } can be obtained from the
knowledge of Probability distribution function
𝑥2
𝑃{𝑥1 < 𝑋 ≤ 𝑥2 } = ∫ 𝑓𝑋 (𝑥 )𝑑𝑥
𝑥1

Proof:
From Probability distribution function
𝑃 {𝑥1 < 𝑋 ≤ 𝑥2 } = 𝐹𝑋 (𝑥2 ) − 𝐹𝑋 (𝑥1 )
𝑥2 𝑥1
= ∫ 𝑓𝑋 (𝑥 )𝑑𝑥 − ∫ 𝑓𝑋 (𝑥 )𝑑𝑥
−∞ −∞
𝑥1 𝑥2 ∞ 𝑥1
= ∫ 𝑓𝑋 (𝑥 )𝑑𝑥 + ∫ 𝑓𝑋 (𝑥 )𝑑𝑥 + ∫ 𝑓𝑋 (𝑥 )𝑑𝑥 − ∫ 𝑓𝑋 (𝑥 )𝑑𝑥
−∞ 𝑥1 𝑥2 −∞
𝑥2
= ∫ 𝑓𝑋 (𝑥 )𝑑𝑥
𝑥1
𝑥2
𝑃{𝑥1 < 𝑋 ≤ 𝑥2 } = ∫ 𝑓𝑋 (𝑥 )𝑑𝑥
𝑥1

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 8

 UNIT-I THE RANDOM VARIABLE

Binomial Density and Distribution Function

This function is used in those applications where only two outcomes are
possible.
Probability density function
𝑁

𝑓𝑋 (𝑥 ) = ∑ 𝑁𝐶𝑘 𝑃𝑘 (1 − 𝑃)𝑁−𝑘 𝛿 (𝑥 − 𝑘 )
𝑘=1

Probability distribution function

𝑁

𝐹𝑋 (𝑥 ) = ∑ 𝑁𝐶𝑘 𝑃𝑘 (1 − 𝑃)𝑁−𝑘 𝑢(𝑥 − 𝑘 )

𝑘=1

where P – Probability of occurrence of event

Poisson Density and Distribution Function

Probability density function
∞
−𝑏
𝑏𝑘
𝑓𝑋 (𝑥 ) = 𝑒 ∑ 𝛿 (𝑥 − 𝑘 )
𝑘!
𝑘=0

where b>0 is a real constant

𝑏=𝜆𝑇
𝜆 = 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑟𝑎𝑡𝑒 𝑇 = 𝑡𝑖𝑚𝑒 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙
Probability distribution function
∞
−𝑏
𝑏𝑘
𝐹𝑋 (𝑥 ) = 𝑒 ∑ 𝑢 (𝑥 − 𝑘 )
𝑘!
𝑘=0

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 9

 UNIT-I THE RANDOM VARIABLE

Applications
It is mostly applied to counting type problems
 The no.of telephone calls made during a period of time
 The no.of defective elements in a given samples
 The no.of items waiting in a queue

Gaussian Density and Distribution Function

This is the most prominently used density function in the field of science and
engineering.
In particular Gaussian function is used to approximate noise in communication
related applications.
Gaussian density function
(𝒙−𝝁𝒙 )𝟐
1 −(
𝟐𝝈𝒙 𝟐
)
𝑓𝑋 (𝑥 ) = 𝑒
√2𝜋𝜎𝑥 2
Where 𝜇𝑥 𝑎𝑛𝑑 𝜎𝑥 2 are mean and variance of random variable X.
𝜎𝑥 is standard deviation.

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 10

 UNIT-I THE RANDOM VARIABLE

Uniform Density and Distribution Function

This function is used to describe the behaviour of the random variable that is same
for a given time interval.
Uniform density function
1
𝑓𝑋 (𝑥 ) = {𝑏 − 𝑎 𝑎 ≤ 𝑋 ≤ 𝑏
0 𝑒𝑙𝑠𝑒 𝑤h𝑒𝑟𝑒
Uniform distribution function
𝑥
𝐹𝑋 (𝑥 ) = ∫ 𝑓𝑋 (𝑥 )𝑑𝑥
−∞
𝑥 1 1 𝑥−𝑎
= ∫ 𝑑𝑥 = 𝑥 ]𝑎𝑥 =
𝑎 𝑏−𝑎 𝑏−𝑎 𝑏−𝑎

𝐹𝑋 (𝑎) = 0
𝑏−𝑎
𝐹𝑋 (𝑏) = =1
𝑏−𝑎
0 𝑥<𝑎
𝑥−𝑎
𝐹𝑋 (𝑥 ) = { 𝑎≤𝑋<𝑏
𝑏−𝑎
1 𝑥≥𝑏

a) Density function b) Distribution function

Applications
 The random distribution of errors introduced in the round off process is
uniformly distributed.
 In digital communications during sampling process.

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 11

 UNIT-I THE RANDOM VARIABLE

Exponential Density and Distribution Function

Probability density function of a random variable X is defined as
1 −(𝑥−𝑎)
𝑓𝑋 (𝑥 ) = { 𝑏 𝑒 𝑥>𝑎
𝑏

0 𝑥<𝑎
where a and b are real constants −∞ < 𝑎 < ∞ 𝑎𝑛𝑑 𝑏 > 0
Probability distribution function
𝑥
𝐹𝑋 (𝑥 ) = ∫ 𝑓𝑋 (𝑥 )𝑑𝑥
−∞
𝑥 1 −(𝑥−𝑎)
𝐹𝑋 (𝑥 ) = ∫ 𝑒 𝑏 𝑑𝑥
−∞ 𝑏
𝑎 𝑥
1 −(𝑥−𝑎) 1 −(𝑥−𝑎)
𝐹𝑋 (𝑥 ) = ∫ 𝑒 𝑏 𝑑𝑥 + ∫ 𝑒 𝑏 𝑑𝑥
−∞ 𝑏 𝑎 𝑏
𝑥−𝑎
−( )
𝐹𝑋 (𝑥 ) = 1 − 𝑒 𝑏

0 𝑥<𝑎
𝑥−𝑎
𝐹𝑋 (𝑥 ) = { 1 − −(
𝑒 𝑏
)
𝑥>𝑎
1 𝑥=∞

Applications
 The distribution of fluctuations in signal strength received by radar
receivers from certain types of targets.
 The distribution of raindrop sizes when a large number of rainstorm
measurements are made.

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 12

 UNIT-I THE RANDOM VARIABLE

Rayleigh Density and Distribution Function

Probability density function of a random variable X is defined as
2 (𝑥−𝑎)2
− 𝑏
( )
𝑓𝑋 (𝑥 ) = 𝑏 𝑥 − 𝑎 𝑒 𝑥≥𝑎
0 𝑥<𝑎
where a and b are real constants −∞ < 𝑎 < ∞ 𝑎𝑛𝑑 𝑏 > 0
Probability distribution function
𝑥
𝐹𝑋 (𝑥 ) = ∫ 𝑓𝑋 (𝑥 )𝑑𝑥
−∞
𝑥 2 (𝑥−𝑎)2
𝐹𝑋 (𝑥 ) = ∫ (𝑥 − 𝑎)𝑒 − 𝑏 𝑑𝑥
−∞ 𝑏
(𝑥−𝑎)2 2
Let =𝑦 ⟹ (𝑥 − 𝑎)𝑑𝑥 = 𝑑𝑦
𝑏 𝑏
𝑥
𝐹𝑋 (𝑥 ) = ∫ 𝑒 −𝑦 𝑑𝑦
𝑎
𝑥
(𝑥−𝑎)2 (𝑥−𝑎)2
− −
= −𝑒 −𝑦 ]𝑎𝑥 = −𝑒 𝑏 ] = 1−𝑒 𝑏
𝑎

0 𝑥<𝑎
(𝑥−𝑎)2
𝐹𝑋 (𝑥 ) = {1 − 𝑒 − 𝑏
𝑥≥𝑎
1 𝑥=∞

a) Density function b) Distribution function

Applications
 It describes the envelope of white noise, when the noise is passed
through a band pass filter.
 Some types of signal fluctuations received by receivers are modelled as
Rayleigh distribution

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 13

 UNIT-I THE RANDOM VARIABLE

OPERATIONS ON RANDOM VARIABLES

 Various operations on random variable will indicate the way it behaves.
 These operations are used to perform statistical analysis of real time
experiment where they exists much randomness.
 Expected values as well as moments are the parameter is used for
describing the behaviour of random variable.

Expected Value or Mean

Expectation is the fundamental operation used in the statistical analysis.
If X is a continuous RV with valid pdf 𝑓𝑋 (𝑥 ), then the expected value of X
or the mean value of X is defined as
∞
𝐸 [𝑋] = 𝑋̅ = ∫ 𝑥 ∗ 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞

If X is a discrete RV
𝑁

𝐸 [𝑋] = 𝑋̅ = ∑ 𝑥𝑖 𝑃(𝑥𝑖 )
𝑖=1

Expected Value of a function of a Random Variable

If X is a continuous RV with valid pdf 𝑓𝑋 (𝑥 ), then the expected value of
function g(x) is defined as
∞
𝐸 [𝑋] = 𝑋̅ = ∫ 𝑔(𝑥 ) 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞

If X is a discrete RV, then the expected value of function g(x) is defined as

𝑁

𝐸 [𝑋] = 𝑋̅ = ∑ 𝑔(𝑥𝑖 ) ( 𝑃(𝑥𝑖 )

𝑖=1

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 14

 UNIT-I THE RANDOM VARIABLE

Properties of Expectations
1. The expected value of a constant is constant.
Proof
∞
𝐸 [𝑋] = ∫ 𝑥. 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞
∞
𝐸 [𝑘 ] = ∫ 𝑘. 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞
∞
𝐸 [𝑋] = 𝑘. ∫ 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞

𝐸 [𝑋] = 𝑘. 1
𝐸 [𝑋] = 𝑘
2. Let 𝐸 [𝑋] be the expected value of a RV ‘X’ then
𝐸 [𝑎𝑋] = 𝑎 𝐸 [𝑋]
Proof
∞
𝐸 [𝑋] = ∫ 𝑥. 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞
∞
𝐸 [𝑎𝑋] = ∫ 𝑎𝑥. 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞
∞
= 𝑎 ∫−∞ 𝑥. 𝑓𝑋 (𝑥 ) 𝑑𝑥

𝐸 [𝑎𝑋] = 𝑎 𝐸 [𝑋]
3. Let 𝐸 [𝑋] be the expected value of a RV ‘X’ then
𝐸 [𝑎𝑋 + 𝑏] = 𝑎𝐸 [𝑋] + 𝑏
∞
𝐸 [𝑋] = ∫ 𝑥. 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞
∞
𝐸 [𝑎𝑋 + 𝑏] = ∫ (𝑎𝑥 + 𝑏). 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞
∞ ∞
= 𝑎 ∫−∞ 𝑥. 𝑓𝑋 (𝑥 ) 𝑑𝑥+ 𝑏. ∫−∞ 𝑓𝑋 (𝑥 ) 𝑑𝑥
𝐸 [𝑎𝑋] = 𝑎𝐸 [𝑋]+b

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 15

 UNIT-I THE RANDOM VARIABLE

Moments:
Moment of a RV describes the deviation from a reference value.
 Moments about the origin
 Moments about the Mean
Moments about the origin
The expected value of a given function 𝑔(𝑥 ) = 𝑋𝑛
is called nth moment about the origin.
𝑚𝑛 = 𝐸 (𝑋𝑛 )
∞
= ∫ 𝑥 𝑛 . 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞
∞
𝑚1 = 𝐸 (𝑋) = ∫ 𝑥. 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞

The first moment about the origin is nothing but mean value (or) expected value
of a random variable.
Second Moment
∞
2)
𝑚2 = 𝐸 ( 𝑋 = ∫ 𝑥 2 . 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞

Moments about the Mean

The expected value of a given function 𝑔(𝑥 ) = (𝑋 − 𝑋̅)𝑛 is called nth
central moment (or) nth moment about the mean.
𝜇𝑛 = 𝐸 (𝑋 − 𝑋̅)𝑛
∞
= ∫−∞(𝑥 − 𝑋̅)𝑛 . 𝑓𝑋 (𝑥 ) 𝑑𝑥
𝜇0 = 1
∞ ∞
𝜇1 = ∫ 𝑥. 𝑓𝑋 (𝑥 ) 𝑑𝑥 − ∫ 𝑋̅. 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞ −∞

= 𝑋̅ − 𝑋̅
=0

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 16

 UNIT-I THE RANDOM VARIABLE

Variance:
The second central moment of a random variable ‘X’ is called Variance.
It is defined as the expected value of a function of the form 𝑔(𝑥 ) = (𝑋 − 𝑋̅)2
where 𝑋̅ − 𝑚𝑒𝑎𝑛 𝑣𝑎𝑙𝑢𝑒.
𝜇2 = 𝜎𝑋 2 = 𝑣𝑎𝑟(𝑥)
The variance is used to calculate the average power of a random signal in
communication related applications.
Relation between variance and moments about the origin
∞

𝜎𝑋 2 = 𝐸 (𝑋 − 𝑋̅)2 = ∫ (𝑥 − 𝑋̅)2 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞
∞

= ∫ (𝑥 2 + 𝑋̅ 2 − 2𝑥𝑋̅) 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞
∞ ∞ ∞

= ∫ 𝑥 2𝑓𝑋 (𝑥 ) 𝑑𝑥 + (𝑋̅)2 ∫ 𝑓𝑋 (𝑥 ) 𝑑𝑥 − 2𝑋̅ ∫ 𝑥 𝑓𝑋 (𝑥 ) 𝑑𝑥

−∞ −∞ −∞

= 𝑚2 + 𝑋̅ 2 − 2 𝑋̅ 𝑋̅
= 𝑚2 + 𝑚12 − 2 𝑚12
𝜎𝑋 2 = 𝑚2 − 𝑚12
Properties of Variance:
1. The variance of a constant is zero.
𝑣𝑎𝑟[𝑘 ] = 0
Proof:
It is known that 𝑣𝑎𝑟[𝑋] = 𝐸 (𝑋 − 𝑋̅)2
2
𝑣𝑎𝑟[𝑘 ] = 𝐸(𝑘 − 𝑘̅ )
as k is constant 𝑘 = 𝑘̅
𝑣𝑎𝑟[𝑘 ] = 𝐸 (𝑘 − 𝑘 )2
𝑣𝑎𝑟[𝑘 ] = 0

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 17

 UNIT-I THE RANDOM VARIABLE

2 Let 𝑣𝑎𝑟 [𝑋] denote the variance of a RV ‘X’ then

𝑣𝑎𝑟[𝑎𝑋] = 𝑎2 𝑣𝑎𝑟[𝑋]
Proof:
𝑣𝑎𝑟[𝑋] = 𝐸 (𝑋 − 𝑋̅)2
𝑣𝑎𝑟[𝑎𝑋] = 𝐸 (𝑎𝑋 − 𝑎𝑋̅)2
= 𝐸 [𝑎2 (𝑋 − 𝑋̅)2 ]
= 𝑎2 𝐸 (𝑋 − 𝑋̅)2
= 𝑎2 𝑣𝑎𝑟[𝑋]
3. Let 𝑣𝑎𝑟 [𝑋] denote the variance of a RV ‘X’ then
𝑣𝑎𝑟 [𝑎𝑋 + 𝑏] = 𝑎2 𝑣𝑎𝑟[𝑋]
Proof:
𝑣𝑎𝑟[𝑋] = 𝐸 (𝑋 − 𝑋̅)2
2
̅̅̅̅̅̅̅̅̅
𝑣𝑎𝑟[𝑎𝑋 + 𝑏] = 𝐸 ((𝑎𝑋 + 𝑏) − (𝑎𝑋 + 𝑏))
2
= 𝐸((𝑎𝑋 + 𝑏) − (𝑎𝑋̅ + 𝑏))
= 𝐸 (𝑎𝑋 − 𝑎𝑋̅)2
= 𝐸 [𝑎2 (𝑋 − 𝑋̅)2 ]
= 𝑎2 𝐸 (𝑋 − 𝑋̅)2
= 𝑎2 𝑣𝑎𝑟[𝑋]
SKEWNESS:
The third central moment of a random variable ‘X’ is called skewness.
It gives the asymmetry of a random variable with reference to a given value.
∞

𝜇3 = 𝐸 (𝑋 − 𝑋̅)3 = ∫ (𝑋 − 𝑋̅)3 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞

It is the degree of distortion from the symmetrical bell curve or the normal
distribution. It measures the lack of symmetry in data distribution.
It differentiates extreme values in one versus the other tail. A symmetrical
distribution will have a skewness of 0.

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 18

 UNIT-I THE RANDOM VARIABLE

There are two types of Skewness: Positive and Negative

Positive Skewness means when the tail on the right side of the distribution is
longer or fatter. The mean and median will be greater than the mode.

Negative Skewness is when the tail of the left side of the distribution is longer
or fatter than the tail on the right side. The mean and median will be less than
the mode.

Note:

 If the skewness is between -0.5 and 0.5, the data are fairly symmetrical.
 If the skewness is between -1 and -0.5(negatively skewed) or between 0.5
and 1(positively skewed), the data are moderately skewed.
 If the skewness is less than -1(negatively skewed) or greater than
1(positively skewed), the data are highly skewed.

Coefficient of skewness:
It is defined as the ratio of 3rd central moment to cube of standard deviation.
𝜇3
= 3
𝜎𝑋

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 19

 UNIT-I THE RANDOM VARIABLE

FUNCTIONS THAT GIVE MOMENTS

Along with probability density function two more functions are also used to
describe the behaviour of a random variable. They are
 CHARACTERISTICS FUNCTION
 MOMENT GENERATING FUNCTIONS
These two functions are useful for calculating the nth moments of a random
variable.
CHARACTERISTICS FUNCTION
Consider a random variable X with pdf 𝑓𝑋 (𝑥 ), then the expected value of the
function 𝑒 𝑗𝜔𝑋 is called characteristics function.
𝜙𝑋 (𝜔) = 𝐸[𝑒 𝑗𝜔𝑋 ]
It is a function of real variable −∞ < 𝜔 < ∞
where j is an imaginary operator
∞
𝜙𝑋 (𝜔) = ∫ 𝑒 𝑗𝜔𝑋 𝑓𝑋 (𝑥 ) dx
−∞

1 ∞ −𝑗𝜔𝑋
𝑓𝑋 (𝑥 ) = ∫ 𝑒 𝜙𝑥 (𝜔) dw
2𝜋 −∞
𝜙𝑥 (𝜔) 𝑎𝑛𝑑 𝑓𝑋 (𝑥 ) are Fourier transform pairs with the sign of the variable is
reversed.

Properties of characteristics function

1. The value of characteristic function is unity at origin, i.e; 𝜔 =0

𝜙𝑋 (0) = 1
Proof
It is known that
𝜙𝑋 (𝜔) = 𝐸[𝑒 𝑗𝜔𝑋 ]

𝜙𝑋 (0) = 𝐸[𝑒 𝑗𝜔0 ] = 𝐸 [1] = 1

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 20
 UNIT-I THE RANDOM VARIABLE

2. The maximum amplitude of characteristics function is unity at origin,

|𝜙𝑋 (𝜔)| ≤ 1
Proof
It is known that
𝜙𝑋 (𝜔) = 𝐸[𝑒 𝑗𝜔𝑋 ]

The amplitude of 𝜙𝑥 (𝜔) is

|𝜙𝑋 (𝜔)| = |𝐸[𝑒 𝑗𝜔𝑋 ]|
∞
= |∫ 𝑒 𝑗𝜔𝑋 𝑓𝑋 (𝑥 ) dx |
−∞

Since |𝑋𝑌| ≤ |𝑋| |𝑌|

∞
|𝜙𝑋 (𝜔)| ≤ ∫ |𝑒 𝑗𝜔𝑋 | | 𝑓𝑋 (𝑥 )| dx
−∞

𝑠𝑖𝑛𝑐𝑒 |𝑒 𝑗𝜔𝑋 | = 1
∞
|𝜙𝑋 (𝜔)| ≤ ∫ | 𝑓𝑋 (𝑥 )| dx
−∞

|𝜙𝑋 (𝜔)| ≤ 1

3. If 𝜙𝑥 (𝜔) be the characteristic function of a random variable X, then

characteristic function of 𝑌 = 𝑎𝑋 + 𝑏 is given by 𝜙𝑌 (𝜔) = 𝑒 𝑗𝜔𝑏 𝜙𝑥 (𝑎𝜔)
Proof:
Given 𝑌 = 𝑎𝑋 + 𝑏
𝜙𝑌 (𝜔) = 𝐸⌈𝑒 𝑗𝜔𝑌 ⌉
= 𝐸⌈𝑒 𝑗𝜔(𝑎𝑋+𝑏)⌉
= 𝐸⌈𝑒 𝑗𝜔𝑎𝑋 𝑒 𝑗𝜔𝑏 ⌉
= 𝑒 𝑗𝜔𝑏 𝐸⌈𝑒 𝑗𝜔𝑎𝑋 ⌉
𝜙𝑌 (𝜔) = 𝑒 𝑗𝜔𝑏 𝜙𝑋 (𝑎𝜔)

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 21

 UNIT-I THE RANDOM VARIABLE

4. If X1 and X2 are two independent random variables then

𝜙𝑋1+𝑋2 (𝜔) = 𝜙𝑋1 (𝜔) ∗ 𝜙𝑋2 (𝜔)
Proof
It is known that
𝜙𝑋 (𝜔) = 𝐸[𝑒 𝑗𝜔𝑋 ]
𝜙𝑋1+𝑋2 (𝜔) = 𝐸[𝑒 𝑗𝜔(𝑋1 +𝑋2) ]
= 𝐸[𝑒 𝑗𝜔𝑋1 𝑒 𝑗𝜔𝑋2 ]
= 𝐸[𝑒 𝑗𝜔𝑋1 ] 𝐸[ 𝑒 𝑗𝜔𝑋2 ]
Since X1 and X2 are two independent random variables
𝜙𝑋1+𝑋2 (𝜔) = 𝜙𝑋1 (𝜔) ∗ 𝜙𝑋2 (𝜔)

5. The nth moment of random variable can be obtained from the knowledge
of characteristic function is
𝑑 𝑛 𝜙𝑋 (𝜔)
𝑚𝑛 = (−𝑗 )𝑛 |
𝑑𝜔 𝑛 𝜔=0

Proof:
Consider 𝜙𝑋 (𝜔) = 𝐸[𝑒 𝑗𝜔𝑋 ]
∞
𝜙𝑋 (𝜔) = ∫ 𝑒 𝑗𝜔𝑥 𝑓𝑋 (𝑥 ) dx
−∞

Differentiate both side w.r.to 𝜔

𝑑 𝑛 𝜙𝑋 (𝜔) 𝑑 𝑛 ∞ 𝑗𝜔𝑥
= ∫ 𝑒 𝑓𝑋 (𝑥 ) dx
𝑑𝜔 𝑛 𝑑𝜔 𝑛 −∞
Changing the order
∞
𝑑 𝑛 𝜙𝑋 (𝜔) 𝑑 𝑛 𝑗𝜔𝑥
=∫ 𝑛𝑒 𝑓𝑋 (𝑥 ) dx
𝑑𝜔 𝑛 −∞ 𝑑𝜔
∞
= ∫ (𝑗𝑥 )𝑛 𝑒 𝑗𝜔𝑥 𝑓𝑋 (𝑥 ) dx
−∞
∞
= (𝑗)𝑛 ∫ (𝑥 )𝑛 𝑒 𝑗𝜔𝑥 𝑓𝑋 (𝑥 ) dx
−∞

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 22

 UNIT-I THE RANDOM VARIABLE

𝑑 𝑛 𝜙𝑋 (𝜔) ∞

𝑛
| = (𝑗) ∫ (𝑥 )𝑛 𝑓𝑋 (𝑥 ) dx
𝑛
𝑑𝜔 𝜔=0 −∞

= (𝑗 ) 𝑛 𝑚 𝑛
𝑑 𝑛 𝜙𝑋 (𝜔)
𝑚𝑛 = (−𝑗 )𝑛 |
𝑑𝜔 𝑛 𝜔=0

MOMENT GENERATING FUNCTIONS

The moment generating function of a random variable is also used to generate
the nth moment about the origin.
Consider a random variable X with pdf 𝑓𝑋 (𝑥 ), then the expected value of the
function 𝑒 𝑣𝑋 is called moment generating function
𝑀𝑋 (𝜈 ) = 𝐸 [𝑒 𝑣𝑋 ]
where 𝜈 is a real variable −∞ < 𝜈 < ∞
∞
𝑀𝑋 (𝜈 ) = ∫ 𝑒 𝑣𝑋 𝑓𝑋 (𝑥 ) dx
−∞

Properties of moment generating function:

1. The moment generating function is unity at origin
𝑀𝑋 (0) = 1
Proof:
∞
𝑀𝑋 (𝜈 ) = ∫ 𝑒 𝑣𝑋 𝑓𝑋 (𝑥 ) dx
−∞
∞
𝑀𝑋 (0) = ∫ 𝑒 0.𝑋 𝑓𝑋 (𝑥 ) dx
−∞
∞
𝑀𝑋 (0) = ∫ 𝑓𝑋 (𝑥 ) dx
−∞

𝑀𝑋 (0) = 1

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 23

 UNIT-I THE RANDOM VARIABLE

2. Let 𝑀𝑋 (𝜈 ) be the moment generating function of a random variable X,

then moment generating function of 𝑌 = 𝑎𝑋 + 𝑏 is given by
𝑀𝑌 (𝜈 ) = 𝑒 𝑣𝑏 𝑀𝑋 (𝜈 )
Proof:
𝑀𝑌 (𝜈 ) = 𝐸 [𝑒 𝑣𝑌 ]
= 𝐸[𝑒 𝑣(𝑎𝑋+𝑏)]
= 𝐸 [𝑒 𝑣𝑎𝑋 𝑒 𝑣𝑏 ]
= 𝑒 𝑣𝑏 𝐸 [𝑒 𝑣𝑎𝑋 ]
𝑀𝑌 (𝜈 ) = 𝑒 𝑣𝑏 𝑀𝑋 (𝜈 )
3. The nth moment of random variable can be obtained from the knowledge
of moment generating function is
𝑑 𝑛 𝑀𝑋 (𝜈 )
𝑚𝑛 = |
𝑑𝑣 𝑛 𝑣=0
Proof:
Differentiate both side w.r.to 𝑣
𝑑 𝑛 𝑀𝑋 (𝜈 ) 𝑑 𝑛 ∞ 𝑣𝑥
= 𝑛 ∫ 𝑒 𝑓𝑋 (𝑥 ) dx
𝑑𝑣 𝑛 𝑑𝑣 −∞
Changing the order
∞𝑑 𝑛 𝑣𝑥
=∫ 𝑛𝑒 𝑓𝑋 (𝑥 ) dx
−∞ 𝑑𝑣
∞
= ∫ (𝑥 )𝑛 𝑒 𝑣𝑥 𝑓𝑋 (𝑥 ) dx
−∞
∞
= ∫ (𝑥 )𝑛 𝑒 𝑣𝑥 𝑓𝑋 (𝑥 ) dx
−∞

𝑑 𝑛 𝑀𝑋 (𝜈 ) ∞

𝑛 | = ∫ (𝑥 )𝑛 𝑓𝑋 (𝑥 ) dx = 𝑚𝑛
𝑑𝑣 𝑣=0 −∞

𝑑 𝑛 𝑀𝑋 (𝜈 )
𝑚𝑛 = |
𝑑𝑣 𝑛 𝑣=0

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 24

 UNIT-I THE RANDOM VARIABLE

TRANSFORMATION ON RANDOM VARIABLES

In majority of the real time applications random variable is to be converted into

another random variable based on certain type of transformation.

Let ‘X’ be a random variable then 𝑌 = 𝑇(𝑋)

where T is the operation (or) transformation performed on ‘X’.

 The nature of Y depends on the type of X as well as type of

transformation is used.
 The random variable X is either continuous (or) discrete (or) mixed.
Based on this number of transmissions exists like linear, non-linear,
segmentation, stair case, etc

The following are the specific cases of transformations

1. Both X and T are continuous and T is either Monotonic Increasing or

Decreasing with X
2. Both X and T are continuous and T is Non-Monotonic
3. X is discrete and T is continuous.

Monotonic Transformation of a Continuous Random Variable

A function is said to be monotonically increasing if 𝑇(𝑋2 ) > 𝑇(𝑋1 ) for

any 𝑥1 < 𝑥2
A function is said to be monotonically decreasing if 𝑇(𝑋2 ) < 𝑇(𝑋1 ) for
any 𝑥1 < 𝑥2

Monotonic Increasing Transformation

For transforming a random variable X into Y the distribution function of both X

and y must be same.

Let there exist a random variable X such that 𝑥 = 𝑥0 then 𝑦0 = 𝑇(𝑥0 ) (or)
𝑥0 = 𝑇 −1(𝑦0 )

As the distribution function should be same

P{𝑌 ≤ 𝑦0 } = P{𝑋 ≤ 𝑥0 }

𝐹𝑌 (𝑦0 ) = 𝐹𝑋 (𝑥0 )
𝑥
We know that 𝐹𝑋 (𝑥 ) = ∫−∞ 𝑓𝑋 (𝑥 )𝑑𝑥

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 25

 UNIT-I THE RANDOM VARIABLE
𝑦0 𝑥0
∫ 𝑓𝑌 (𝑦)𝑑𝑦 = ∫ 𝑓𝑋 (𝑥 )𝑑𝑥
−∞ −∞

Differentiating on both sides w.r.to ‘y’

𝑑 𝑦0 𝑑 𝑥0
(
∫ 𝑓 𝑦 𝑑𝑦 =) ∫ 𝑓 (𝑥 )𝑑𝑥
𝑑𝑦 −∞ 𝑌 𝑑𝑦 −∞ 𝑋
Using Leibniz rule
𝑑𝑥0
𝑓𝑌 (𝑦0 ) = 𝑓𝑋 (𝑥0 )
𝑑𝑦
𝑑
𝑓𝑌 (𝑦0 ) = 𝑓𝑋 (𝑇 −1(𝑦0 )) (𝑇 −1(𝑦0 ))
𝑑𝑦
The above transformation is applicable for every value of y and x resulting in
𝑑𝑥
𝑓𝑌 (𝑦) = 𝑓𝑋 (𝑥 )
𝑑𝑦

Monotonic Decreasing Transformation

{𝑌 ≤ 𝑦0 } = {𝑋 ≥ 𝑥0 }
P{𝑌 ≤ 𝑦0 } = P{𝑋 ≥ 𝑥0 }
P{𝑌 ≤ 𝑦0 } = 1 − P{𝑋 < 𝑥0 }
𝐹𝑌 (𝑦0 ) = 1 − 𝐹𝑋 (𝑥0 )
𝑥
We know that 𝐹𝑋 (𝑥 ) = ∫−∞ 𝑓𝑋 (𝑥 )𝑑𝑥
𝑦0 𝑥0
∫ 𝑓𝑌 (𝑦)𝑑𝑦 = 1 − ∫ 𝑓𝑋 (𝑥 )𝑑𝑥
−∞ −∞

Differentiating on both sides w.r.to ‘y’

𝑑 𝑦0 𝑑 𝑥0
∫ 𝑓 (𝑦)𝑑𝑦 = [1 − ∫ 𝑓𝑋 (𝑥 )𝑑𝑥 ]
𝑑𝑦 −∞ 𝑌 𝑑𝑦 −∞

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 26

 UNIT-I THE RANDOM VARIABLE

Using Leibniz rule

𝑑𝑥0
𝑓𝑌 (𝑦0 ) = −𝑓𝑋 (𝑥0 )
𝑑𝑦
𝑑
𝑓𝑌 (𝑦0 ) = −𝑓𝑋 (𝑇 −1(𝑦0 )) (𝑇 −1(𝑦0 ))
𝑑𝑦
The above transformation is applicable for every value of y and x resulting in
𝑑𝑥
𝑓𝑌 (𝑦) = −𝑓𝑋 (𝑥 )
𝑑𝑦
For monotonic transformation either increasing or decreasing, the density
function of y is
𝑑𝑥
𝑓𝑌 (𝑦) = 𝑓𝑋 (𝑥 ) | |
𝑑𝑦

Non-Monotonic Transformation of a Continuous Random Variable

Consider a random variable Y is a Non-Monotonic Transformation of a

Continuous Random Variable X as shown in figure

For a given event {𝑌 ≤ 𝑦0 }, there are more than one value of X.

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 27

 UNIT-I THE RANDOM VARIABLE

It is observed that the event {𝑌 ≤ 𝑦0 } corresponds to the events {𝑋 ≤ 𝑥1 𝑎𝑛𝑑 𝑥2 < 𝑋 ≤

𝑥3 }

Thus the probability of the event {𝑌 ≤ 𝑦0 } is equal to the probability of the

event {𝑋/𝑌 ≤ 𝑦0 }

P{𝑌 ≤ 𝑦0 } = P{𝑋/𝑌 ≤ 𝑦0 }

𝐹𝑌 (𝑦0 ) = ∫ 𝑓𝑋 (𝑥 )𝑑𝑥
𝑋/𝑌≤𝑦0

By differentiating, the density function is given by

𝑑 𝑑
𝐹𝑌 (𝑦0 ) = ∫ 𝑓 (𝑥 )𝑑𝑥
𝑑𝑦 𝑑𝑦 𝑋/𝑌≤𝑦0 𝑋

𝑓𝑋 (𝑥𝑛 )
𝑓𝑌 (𝑦) = ∑
𝑑𝑇(𝑥)
𝑛 | |
𝑑𝑥
𝑑𝑥1 𝑑𝑥2 𝑑𝑥3
𝑓𝑌 (𝑦) = 𝑓𝑋 (𝑥1 ) | | + 𝑓𝑋 (𝑥2 ) | | + 𝑓𝑋 (𝑥3 ) | | ….
𝑑𝑦 𝑑𝑦 𝑑𝑦

MEAN AND VARIANCE FOR POISSON RANDOM VARIABLE

We know that the Probability density function for Poisson random variable is
∞
−𝜆
𝜆𝑘
𝑓𝑋 (𝑥 ) = 𝑒 ∑ 𝛿 (𝑥 − 𝑘 )
𝑘!
𝑘=0

The probability function is

𝜆𝑥
𝑃 (𝑥 ) = 𝑒 −𝜆 𝑥 = 0,1,2,3 … …
𝑥!
Expected Value

If X is a discrete RV
𝑁

𝐸 [𝑋] = 𝑋̅ = ∑ 𝑥𝑖 𝑃(𝑥𝑖 )
𝑖=1

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 28

 UNIT-I THE RANDOM VARIABLE
∞
𝜆𝑥
𝐸 [𝑋] = 𝑋̅ = ∑ 𝑥 𝑒 −𝜆
𝑥!
𝑥=0
∞
−𝜆
𝜆𝑥
= ∑𝑥𝑒
𝑥 (𝑥 − 1)!
𝑥=0
∞
−𝜆
𝜆𝑥
= ∑𝑥𝑒
𝑥 (𝑥 − 1)!
𝑥=0
∞
−𝜆
𝜆 𝜆𝑥−1
=∑ 𝑒
(𝑥 − 1)!
𝑥=1
∞
−𝜆
𝜆𝑥−1
=𝜆𝑒 ∑
(𝑥 − 1)!
𝑥=1

−𝜆
𝜆0 𝜆1 𝜆2
=𝜆𝑒 ( + + +⋯.)
0! 1! 2!
= 𝜆 𝑒 −𝜆 𝑒 𝜆
𝐸 [𝑋] = 𝑚1 = 𝜆
VARIANCE:

𝜎𝑋 2 = 𝑚2 − 𝑚12
𝑁

𝐸 (𝑋2 ) = 𝑚2 = ∑ 𝑥𝑖 2 𝑃(𝑥𝑖 )
𝑖=1
∞
2 −𝜆
𝜆𝑥
𝑚2 = ∑ 𝑥 𝑒
𝑥!
𝑥=0

𝑥 2 = 𝑥 (𝑥 − 1) + 𝑥
∞
−𝜆
𝜆𝑥
= ∑ (𝑥 (𝑥 − 1) + 𝑥 ) 𝑒
𝑥!
𝑥=0
∞ ∞
−𝜆
𝜆𝑥 −𝜆
𝜆𝑥
( )
= ∑𝑥 𝑥−1 𝑒 +∑𝑥𝑒
𝑥! 𝑥!
𝑥=0 𝑥=0

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 29

 UNIT-I THE RANDOM VARIABLE
∞ ∞
−𝜆
𝜆𝑥 −𝜆
𝜆𝑥
= ∑ 𝑥 (𝑥 − 1) 𝑒 +∑𝑥𝑒
𝑥 (𝑥 − 1)(𝑥 − 2)! 𝑥 (𝑥 − 1)!
𝑥=0 𝑥=0
∞ ∞
−𝜆
𝜆2 𝜆𝑥−2 −𝜆
𝜆 𝜆𝑥−1
=∑ 𝑒 +∑ 𝑒
(𝑥 − 2)! (𝑥 − 1)!
𝑥=2 𝑥=1
∞ ∞
2 −𝜆
𝜆𝑥−2 −𝜆
𝜆𝑥−1
=𝜆 𝑒 ∑ +𝜆𝑒 ∑
(𝑥 − 2)! (𝑥 − 1)!
𝑥=2 𝑥=1

2 −𝜆
𝜆0 𝜆1 𝜆2 −𝜆
𝜆0 𝜆1 𝜆2
=𝜆 𝑒 ( + + +⋯.)+ 𝜆 𝑒 ( + + +⋯.)
0! 1! 2! 0! 1! 2!
= 𝜆2 𝑒 −𝜆 𝑒 𝜆 + 𝜆 𝑒 −𝜆 𝑒 𝜆
= 𝜆2 + 𝜆
Variance
𝜎 2 = 𝑚2 − 𝑚12
𝜎 2 = 𝜆2 + 𝜆 − 𝜆2
𝜎2 = 𝜆
For the Poisson random variable
𝐸 [𝑋] = 𝝈𝟐 = 𝜆

MEAN AND VARIANCE FOR UNIFORM RANDOM VARIABLE

We know that the Probability density function for uniform random variable is
1
𝑓𝑋 (𝑥 ) = 𝑎≤𝑋≤𝑏
𝑏−𝑎
=0 𝑜𝑡ℎ𝑒𝑤𝑖𝑠𝑒
Expected Value
∞
𝑚1 = 𝐸 [𝑋] = 𝑋̅ = ∫ 𝑥 ∗ 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞
𝑏 1
= ∫ 𝑥 𝑑𝑥
𝑎 𝑏−𝑎
𝑏
1
= ∫ 𝑥 𝑑𝑥
𝑏−𝑎 𝑎

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 30

 UNIT-I THE RANDOM VARIABLE
𝑏
1 𝑥2
= ]
𝑏−𝑎 2 𝑎

1 𝑏 2 − 𝑎2
=
2 𝑏−𝑎
1 (𝑏 − 𝑎)(𝑏 + 𝑎)
=
2 𝑏−𝑎
(𝑏 + 𝑎)
𝑚1 = 𝐸 [𝑋] =
2
VARIANCE:

𝜎𝑋 2 = 𝑚2 − 𝑚12
∞
2)
𝑚2 = 𝐸 ( 𝑋 = ∫ 𝑥 2 . 𝑓𝑋 (𝑥 ) 𝑑𝑥
−∞
𝑏
1
= ∫ 𝑥2 𝑑𝑥
𝑎 𝑏−𝑎
𝑏
1
= ∫ 𝑥 2 𝑑𝑥
𝑏−𝑎 𝑎
𝑏
1 𝑥3
= ]
𝑏−𝑎 3 𝑎

1 𝑏 3 − 𝑎3
=
3 𝑏−𝑎
1 (𝑏 − 𝑎)(𝑏2 + 𝑎2 + 𝑎𝑏)
=
3 𝑏−𝑎
(𝑏 2 + 𝑎2 + 𝑎𝑏)
𝑚2 =
3
Variance

𝜎 2 = 𝑚2 − 𝑚12

2
(𝑏2 + 𝑎2 + 𝑎𝑏) 𝑏+𝑎 2
𝜎 = −( )
3 2

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 31

 UNIT-I THE RANDOM VARIABLE

2
(𝑏2 + 𝑎2 + 𝑎𝑏) (𝑏2 + 𝑎2 + 2𝑎𝑏)
𝜎 = −
3 4
2
(𝑏 2 + 𝑎2 − 2𝑎𝑏)
𝜎 =
12
2
(𝑏 − 𝑎)2
𝜎 =
12
For the Uniform random variable
(𝑏 + 𝑎)
𝐸 [𝑋] =
2
2
(𝑏 − 𝑎)2
𝜎 =
12
PROBLEMS
1. Find the constant ‘b’ so that the given density function is a valid function.
3𝑥⁄
𝑓𝑋 (𝑥 ) = {𝑒
4, 0 < 𝑥 < 𝑏
0, 𝑒𝑙𝑠𝑒 𝑤ℎ𝑒𝑟𝑒
Sol:
∞
𝑤𝑒 𝑘𝑛𝑜𝑤 𝑡ℎ𝑎𝑡 ∫ 𝑓𝑋 (𝑥 )𝑑𝑥 = 1
−∞

𝑏
3𝑥⁄
∫ 𝑒 4 𝑑𝑥 =1
0
𝑏
3𝑥
𝑒 ⁄4
[ ] =1
3
4 0

3𝑏 3
[𝑒 4 − 1] =
4

3𝑏 7
𝑒4 =
4

3𝑏 7
= ln ( )
4 4
b= 0.7461

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 32

 UNIT-I THE RANDOM VARIABLE

2. Assume automobile arrivals at a gasoline station are Poisson and occurs

at an average rate of 50per/Hour. The station has only one gasoline pump.
If all cars are assumed to require one minute to obtain fuel, What is the
probability that a weighting line will occur at the pump?

Sol:
If two or more cars arrive, then weighting line will occur at the pump

𝑃(𝑋 ≥ 2) = 1 − 𝑃 (𝑋 ≤ 1)

= 1 − 𝐹𝑋 (1)
∞
−𝑏
𝑏𝑘
Poisson distribution 𝐹𝑋 (𝑥 ) = 𝑒 ∑
𝑘!
𝑘=0
1
−𝑏
𝑏𝑘
= 1−𝑒 ∑
𝑘!
𝑘=0
50
𝑏=𝜆𝑇= ∗ 1 = 5/6
60
1 5 𝑘
( )
= 1−𝑒 −5/6
∑ 6
𝑘!
𝑘=0
5 5
= 1 − 𝑒 −6 ∗
6
𝑃(𝑋 ≥ 2) = 0.203

3. A RV ‘X’ is having probability distribution function

𝑛2
𝐹𝑋 (𝑥 ) = ∑12
𝑛=1 𝜇(𝑥 − 𝑛).
650

Obtain a) 𝑃 (−∞ < 𝑋 ≤ 6) b) 𝑃(𝑋 > 4) c) 𝑃(6 < 𝑋 ≤ 9)

𝑛2
Sol: 𝑃 (−∞ < 𝑋 ≤ 6) = ∑6𝑛=1
650
1
= ∑6 𝑛2
650 𝑛=1

1 𝑛(𝑛+1)(𝑛+2)
= [ ]
650 6 6

= 0.14

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 33

 UNIT-I THE RANDOM VARIABLE

𝑏)𝑃 (𝑋 > 4) = 1 − 𝑃 (𝑋 ≤ 4)

𝑛2
= 1 − ∑4𝑛=1 = 0.9538
650

𝑃 (6 < 𝑋 ≤ 9)

𝑃{𝑥1 < 𝑋 ≤ 𝑥2 } = 𝐹𝑋 (𝑥2 ) − 𝐹𝑋 (𝑥1)

= 𝐹𝑋 (9) − 𝐹𝑋 (6)

𝑛2 𝑛2
= ∑9𝑛=1 − ∑6𝑛=1 = 0.2984
650 650

4. A random variable X has the density function is given. Find k, FX(x)

and P(X≥0)
  1 
k   x 
f X ( x)    1  x 2 
,

 0, elsewhere
∞
Sol: We know that ∫−∞ 𝑓𝑋 (𝑥 )𝑑𝑥 = 1
∞ 1
𝑘∫ 2 𝑑𝑥 = 1
−∞ 1 + 𝑥

𝑘 tan−1 𝑥 ]∞
−∞ = 1

𝑘 (tan−1(∞) − tan−1(−∞)) = 1

𝜋 𝜋 1
𝑘( + ) = 1 ⟹ 𝑘 =
2 2 𝜋
ii) The distribution function FX(x)
𝑥 1
𝐹𝑋 (𝑥 ) = 𝑘 ∫ 2 𝑑𝑥
−∞ 1 + 𝑥
= 𝑘 tan−1𝑥 ]−∞
𝑥

= 𝑘 (tan−1(𝑥 ) − tan−1 (−∞))

1 𝜋
= (tan−1 (𝑥 ) + )
𝜋 2

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 34

 UNIT-I THE RANDOM VARIABLE
∞ 1
iii)𝑃(𝑋 ≥ 0) = 𝑘 ∫0 𝑑𝑥
1+𝑥 2

= 𝑘 tan−1𝑥 ]∞
0

= 𝑘 (tan−1(∞) − tan−1(0))

1 𝜋
= ( ) = 0.5
𝜋 2

5. A discrete random variable X takes values from 1 to 5 with

probabilities P(X) as given below. Compute the mean and
variance of the random variable X.

X 1 2 3 4 5
P(X) 0.1 0.2 0.4 0.2 0.1

Sol:
𝑁

𝑚𝑒𝑎𝑛 𝑚1 = 𝐸 [𝑋] = 𝑋̅ = ∑ 𝑥𝑖 𝑃 (𝑥𝑖 )

𝑖=1
5

= ∑ 𝑥𝑖 𝑃 (𝑥𝑖 )
𝑖=1

𝑋̅ = 1 ∗ (0.1) + 2 ∗ (0.2) + 3 ∗ (0.3) + 4 ∗ (0.2) + 5 ∗ (0.1) = 𝟑

𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒 𝜎𝑋 2 = 𝑚2 − 𝑚1 2
𝑁

𝑚2 = ∑ 𝑥𝑖 2 𝑃 (𝑥𝑖 )
𝑖=1

𝑚2 = 1 ∗ (0.1) + 4 ∗ (0.2) + 9 ∗ (0.3) + 16 ∗ (0.2) + 25 ∗ (0.1)

𝑚2 = 9.3
𝜎𝑋 2 = 𝑚2 − 𝑚12
= 9.3 − 9
= 𝟎. 𝟑
A random variable X has X  3, X 2  11, and  x2  2. For a new
6.
random variable Y= 2X-3, find

𝑖) 𝑌̅ 𝑖𝑖) ̅̅̅̅
𝑌2 𝑖𝑖𝑖) 𝜎𝑌 2

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 35

 UNIT-I THE RANDOM VARIABLE

Sol: Given Y= 2X-3

𝑌̅ = 𝐸 [2𝑋 − 3]

= 2 𝐸 [𝑋] − 3

= 2 (−3) − 3 = −9

̅𝑌̅̅2̅ = 𝐸 [(2𝑋 − 3)2 ]

= 𝐸 [4𝑋2 − 12𝑋 + 9]
= 4𝐸 [𝑋2 ] − 12𝐸 [𝑋] + 9
= (4 ∗ 11) − (12 ∗ −3) + 9 = 89
𝜎𝑌 2 = 𝑚2 − 𝑚12
𝜎𝑌 2 = 89 − 81 = 8
(OR)
𝜎𝑌 2 = 𝑣𝑎𝑟(𝑌)
= 𝑣𝑎𝑟(2𝑋 − 3)
Since the variance of a constant is zero
𝑣𝑎𝑟[𝑎𝑋] = 𝑎2 𝑣𝑎𝑟[𝑋]
= 4𝑣𝑎𝑟(𝑋)

=4∗2 =8

7. Obtain (i)Moment generating Function (ii) E[X] of the

probability density function of a random variable X is given
as
𝑓𝑋 (𝑥 ) = 𝑎𝑒 −𝑥 for x ≥ 0.
Consider
∞
𝑣𝑋 ]
𝑀𝑋 (𝜈 ) = 𝐸 [𝑒 = ∫ 𝑒 𝑣𝑥 𝑎 𝑒 −𝑥 dx
0
∞
∞
−(1−𝑣)𝑥
𝑒 −(1−𝑣)𝑥
= 𝑎∫ 𝑒 dx = ]
0 1−𝑣 0

1
𝑀𝑋 (𝜈 ) =
1−𝑣

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 36

 UNIT-I THE RANDOM VARIABLE

𝑑 𝑛 𝑀𝑋 (𝜈 )
𝑤𝑒 𝑘𝑛𝑜𝑤 𝑡ℎ𝑎𝑡 𝑚𝑛 = |
𝑑𝑣 𝑛 𝑣=0
𝑑𝑀𝑋 (𝜈 )
𝑚1 = |
𝑑𝑣 𝑣=0
1
= | =1
(1 − 𝑣 )2 𝑣=0
8. Find the characteristic function of a uniformly distributed random
variable X in the range[0,1] and hence find out 𝑚1
Sol:
We know that the Probability density function for uniform random
variable is
1
𝑓𝑋 (𝑥 ) = 𝑎≤𝑋≤𝑏
𝑏−𝑎
=0 𝑜𝑡ℎ𝑒𝑤𝑖𝑠𝑒
in the range[0,1] a=0, b=1
𝑓𝑋 (𝑥 ) = 1 0 ≤ 𝑋 ≤ 1
=0 𝑜𝑡ℎ𝑒𝑤𝑖𝑠𝑒
The characteristic function is
∞
𝜙𝑋 (𝜔) = ∫ 𝑒 𝑗𝜔𝑋 𝑓𝑋 (𝑥 ) dx
−∞
1 𝑒 𝑗𝜔𝑋
=∫ dx
0 𝑗𝜔
1
𝑒 𝑗𝜔𝑋
= ]
𝑗𝜔 0
1 𝑗𝜔
𝜙𝑋 (𝜔) = [𝑒 − 1 ]
𝑗𝜔
To find out 𝑚1
𝑑 𝑛 𝜙𝑋 (𝜔)
𝑚𝑛 = (−𝑗 )𝑛 |
𝑑𝜔 𝑛 𝜔=0
𝑑𝜙𝑋 (𝜔)
𝑚1 = (−𝑗) |
𝑑𝜔 𝜔=0
𝑑𝜙𝑋 (𝜔) 𝑑 𝑒 𝑗𝜔 − 1
= ⌈ ⌉
𝑑𝜔 𝑑𝜔 𝜔
1
=
2

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 37

 UNIT-I THE RANDOM VARIABLE

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 38

 UNIT-I THE RANDOM VARIABLE

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 39

 UNIT-I THE RANDOM VARIABLE

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 40

 UNIT-I THE RANDOM VARIABLE

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 41

 UNIT-I THE RANDOM VARIABLE

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 42

 UNIT-I THE RANDOM VARIABLE

CH SIVA RAMA KRISHNA Dept. of ECE, LBRCE Page 43