Module: 2 Random Variables

• Introduction -random variables- probability mass Function,

distribution and density functions - joint Probability distribution and
joint density functions- Marginal, conditional distribution and density
functions- Mathematical expectation, and its properties Covariance,
moment generating function – characteristic function.
1. Random experiment: It is a process or procedure that produces a set of outcomes, where
the outcome is uncertain and cannot be predicted with certainty in advance. Example:
Throwing a die, a pack of cards.
2. Outcome: The result of a random experiment will be called an outcome.
3. Trial: A trial refers to any single performance of a random experiment. In each trial, one
of the possible outcomes occurs. Example: Tossing of a coin: {H, T}
4. Event: An event is a specific set of outcomes of a random experiment. An event can
consist of one or more outcomes. Events are often what we are interested in when
conducting an experiment.
5. Exhaustive events or cases: The complete set of all possible outcomes in a random
experiment. In other words, these events cover every possible outcome that could occur
in a trial.
6. Favourable events or cases: The number of cases favourable to an event in a trail is the
number of outcomes which entail the happening of the event.
Example: In drawing a card from a pack of cards the number of cases favourable to
drawing of an ace is 4, for drawing a spade card is 13 and for drawing a red card is 26.
Random Variable
• Intuitively by a random variable (r.v) means a real number X connected
with the outcome of a random experiment E.
• Let S be the sample space associated with a given random experiment E .
A real-valued function defined on 𝑺 and taking values in 𝑹(−∞, ∞) is
called a one-dimensional random variable.
If the function values are ordered pairs of real numbers (i.e., vectors in two-
space), the function is said to be a two-dimensional random variable.
More generally, an n-dimensional random variable is simply a function
whose domain is S and whose range is a collection of n-tuples of real
numbers (vectors in n-space).
Let us consider the probability space, the triplet 𝑺, 𝜝, 𝜬 , where S is the sample space,
viz, space of outcomes, 𝜝 is the 𝝈- field of subsets in S and 𝜬 is a probability function on
𝜝.
Definition: A random variable is a function 𝑋(𝝎)with domain S and range (−∞, ∞) such
that for every real number a, the event 𝜔 ∶ 𝑋 𝝎 ≤ 𝑎 ∈ 𝜝.

Example: Suppose Two (unbiased) coins are tossed 𝑋 = 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 ℎ𝑒𝑎𝑑𝑠.

𝟑
The sample space S = {HH, HT, TH, TT} 𝑷 𝑋 ≤ 1 = 𝑷{𝐻𝐻, 𝐻𝑇, 𝑇𝐻} =
𝟒
Note: One-dimensional r.v. will be denoted by capital letters, X, Y, Z….etc. A typical
outcome of the experiment will be denoted by 𝝎 or e.
The values which X, Y, Z….etc, can assume are denoted by lower case letters.,
x, y, z…etc.
Discrete Random Variable
If 𝑋 is a random variable which can take a finite number or countably infinite number of
values, then 𝑋 is called a discrete random variable.
Example: Marks obtained in a test, number of accidents per month, number of telephone calls
per unit time, number of successes in n trails and so on.
Probability Mass function: If 𝑋 is a discrete random variable with distinct values
𝑥1 , 𝑥2 , … … , 𝑥𝑛 then the function 𝒑𝑿 (𝒙)is defined as :
 P ( X = xi ) = pi if x = xi
p X ( xi ) = 
0 if x  xi i = 0,1, 2,...n
is called the probability mass function or probability function of random variable 𝑿.
The numbers 𝑝 𝑥𝑖 ; 𝑖 = 1, 2, … . . must satisfy the following conditions:
(i) 𝑝 𝑥𝑖 ≥ 0 ∀ 𝑖
(ii) σ𝑛𝑖=1 𝑝 𝑥𝑖 = 1
Distribution Function
Definition: Let 𝑿 be a random variable. The function 𝑭 defined for all
real 𝒙 by
𝐹𝑋 𝑥 = 𝐹 𝑥 = 𝑃 𝑋 ≤ 𝑥 = 𝑃 𝜔 ∶ 𝑋 𝜔 ≤ 𝑥 , −∞ < 𝑥 < ∞
is called the distribution function or cumulative distribution function of r.v. (𝑿).

The domain of the distribution function is −∞, ∞ and its range is [0,1]
Properties:
If 𝒙 is a real number, the set of all 𝝎 in S such that 𝐗 𝝎 = 𝒙 is,denoted by 𝑿 = 𝒙.
1. 𝑃 𝑋=𝑥 =𝑃 𝜔∶𝑋 𝜔 =𝑥

2. 𝑃 𝑋 ≤ 𝑎 = 𝑃 𝜔 ∶ 𝑋 𝜔 ∈ −∞, 𝑎

3. 𝑃 𝑎 < 𝑋 ≤ 𝑏 = 𝑃 𝜔 ∶ 𝑋 𝜔 ∈ 𝑎, 𝑏 = 𝐹 𝑏 − 𝐹(𝑎)

𝑃 𝑎 ≤ 𝑋 ≤ 𝑏 = 𝑃(𝑋 = 𝑎) + [𝐹 𝑏 − 𝐹(𝑎)]
𝑃 𝑎 < 𝑋 < 𝑏 = 𝐹 𝑏 − 𝐹(𝑎) − 𝑃(𝑋 = 𝑏)
𝑃 𝑎 ≤ 𝑋 < 𝑏 = 𝐹 𝑏 − 𝐹(𝑎) − 𝑃(𝑋 = 𝑏) + 𝑃(𝑋 = 𝑎)

4. 𝑃 𝑋 = 𝑎 𝑜𝑟 𝑋 = 𝑏 = 𝑃 𝑋 = 𝑎 ∪ 𝑋 = 𝑏

5. 𝑃 𝑋 = 𝑎 𝑎𝑛𝑑 𝑋 = 𝑏 = 𝑃{(𝑋 = 𝑎) ∩ (𝑋 = 𝑏)}

Properties
If 𝑭 is the distribution function of the r.v. 𝑿 and if 𝑎 < 𝑏, then
6. 𝑃 𝑎 < 𝑋 ≤ 𝑏 = 𝑃 𝑋 ≤ 𝑏 − 𝑃 𝑋 ≤ 𝑎 = 𝐹 𝑏 − 𝐹 𝑎

7. (i) 0 ≤ 𝐹 𝑥 ≤ 1
(ii) 𝐹 𝑥 ≤ 𝐹 𝑦 if 𝑥 < 𝑦

8. 𝐹 −∞ = lim 𝐹 𝑥 = 0 and
𝑥→−∞

𝐹 ∞ = lim 𝐹 𝑥 = 1
𝑥→∞
Discrete Distribution Function:
A countable number of points 𝑥1 , 𝑥2 , 𝑥3 , … . . 𝑥𝑛
𝑝 𝑥𝑖 ≥ 0 ∀ 𝑖,

σ𝑛𝑖=1 𝑝 𝑥𝑖 = 1 such that

𝐹 𝑥 = ෍ 𝑝𝑖 = ෍ 𝑃(𝑋 = 𝑥𝑖 )
𝑖: 𝑥𝑖 ≤𝑥 𝑥𝑖

If 𝑥𝑖 is just integer 𝑖, so that 𝑃 𝑋 = 𝑖 = 𝑝𝑖 ; 𝑖 = 1, 2, 3, … . . Then 𝑭(𝒙) is a “step

function” having jump 𝒑 at 𝑖 and being constant between each pair of integers.

Properties:
1. 𝑝 𝑥𝑗 = 𝑃 𝑋 = 𝑥𝑗 = 𝐹 𝑥𝑗 − 𝐹 𝑥𝑗−1 , where F is the d.f. of 𝑋.
From a lot of 10 items containing 3 defectives, a sample of 4 items is drawn at
random. Let the random variable 𝑋 denote the number of defective items in the
sample. Answer the following when the sample is drawn without replacement.
(i) Find the probability distribution of X.
(ii) Find P (X ≤ 1), P (X < 1) and P (0 < i < 2)
 D  N − D 
 k  n − k 
Solution: P ( X = k ) =   
N
n 
 
where D = number of defective item
N = total number of items
n= number of items drawn in the sample
k = the number of defective items in the sample
i)  3  7  7 ii)
 0  4  1  4 
P( X = 0 ) =    =   = 1 35 = 35 = 1
1 1 2
 10   10  210 210 6 P ( X  1) = P ( X = 0 ) + P ( X = 1) = + =
4  4  6 2 3
   
1
 3  7  7 P ( X  1) = P ( X = 0 ) =
 1  3  3   3 
6
P ( X = 1) =    =   = 3  35 = 105 = 1 1
P ( 0  X  2 ) = P( X = 1) =
 10   10  210 210 2 2
4  4 
   
 3  7  7
 2  2  3   2 
P ( X = 2 ) =    =   = 3  21 = 63 = 3
 10   10  210 210 10
4  4 
   
 3  7  7
 3  1  1  1 
     1 7 7 1
P ( X = 3) = = = = =
 
10  
10 210 210 30
4  4 
   
A random variable 𝑋 has the following probability function:

Values of
0 1 2 3 4 5 6 7
𝑿=𝒙
𝒑(𝑿 = 𝒙) 0 𝑘 2𝑘 2𝑘 3𝑘 𝑘2 2𝑘 2 7𝑘 2 +k

(i) Find 𝑘, (ii) Evaluate 𝑃 𝑋 < 6 , 𝑃 𝑋 ≥ 6 𝑎𝑛𝑑 𝑃 0 < 𝑋 < 5

1
(iii) If 𝑃 𝑋 ≤ 𝑐 > find the minimum value of c,
2
(iv) Determine the distribution function of 𝑋.
 7
Solution : i) Since p ( x) = 1
x =0

k + 2k + 2k + 3k + k 2 + 2k 2 + 7k 2 + k = 1
10k 2 + 9k − 1 = 0
1
k = −1,
10
ii ) P(X  6) = P ( X = 0 ) + P ( X = 1) + ... + P ( X = 5 )
1 2 2 3 1 81
= + + + + =
10 10 10 10 100 100
19
P ( X  6) = 1 − P ( X  6) =
100
4
P ( 0  X  5 ) = P( X = 1) + P( X = 2) + P( X = 3) + P( X = 4) = 8k =
5
 1
iii) P( X  c)   c = 4
2

𝑋 𝑭𝑿 𝒙 = 𝑷(𝑿 ≤ 𝒙)
0 0
iv) 1 1/10
2 3/10
3 5/10
4 4/5
5 81/100
6 83/100
7 1
Continuous Random Variable
Definition: A random variable 𝑋 is said to be continuous if it can take
all possible values between certain limits.

A continuous random variable is a r.v. that can be measured to any

desired degree of accuracy.

Example: Age, height, weight etc.

Probability Density Function (p.d.f)
Definition: Consider the small interval (𝒙, 𝒙 + 𝒅𝒙)of length 𝒅𝒙 round the
point 𝒙. Let 𝒇(𝒙) be any continuous function of 𝑥 so that 𝒇 𝒙 𝒅𝒙 represents
the probability that 𝑋 falls in the infinitesimal interval (𝒙, 𝒙 + 𝒅𝒙).
𝑃 𝑥 ≤ 𝑋 ≤ 𝑥 + 𝑑𝑥 = 𝑓𝑋 𝑥 𝑑𝑥

𝑃 𝑥 ≤ 𝑋 ≤ 𝑥 + 𝛿𝑥
𝑓𝑋 𝑥 = lim
𝛿𝑥→0 𝛿𝑥
The probability for a variate value to line in the interval 𝑑𝑥 is 𝑓(𝑥)𝑑𝑥 and
hence the probability for a variate value to fall in the interval [𝛼, 𝛽] is:
𝛽
𝑃 𝛼 ≤ 𝑋 ≤ 𝛽 = න 𝑓 𝑥 𝑑𝑥
𝛼
Continuous Distribution Function

Definition: If 𝑋 is continuous r.v. with the p.d.f.𝑓(𝑥), then the function

𝑥
𝐹𝑋 𝑥 = 𝑃 𝑋 ≤ 𝑥 = ‫׬‬−∞ 𝑓 𝑡 𝑑𝑡 ; −∞ < 𝑥 < ∞ is called the distribution
function (d.f.) or sometimes the cumulative distribution function (c.d.f.) of the
random variable 𝑋.
Properties of 𝒑. 𝒅. 𝒇. :
(i) 𝑓 𝑥 ≥ 0
∞
(ii) ‫׬‬−∞ 𝑓 𝑥 𝑑𝑥 = 1

(iii) The probability 𝑃(𝐸)given by: ‫ 𝑥𝑑 𝑥 𝑓 𝐸׬‬is well defined for any event 𝐸.

(iv) 𝑃 𝑋 = 𝑐 = 0, ∀ 𝑐
𝑃(𝛼 ≤ 𝑋 ≤ 𝛽) = 𝑃(𝛼 ≤ 𝑋 < 𝛽) = 𝑃(𝛼 < 𝑋 ≤ 𝛽) = 𝑃(𝛼 < 𝑋 < 𝛽)
Properties: 1. 0 ≤ 𝐹(𝑥) ≤ 1 ; −∞ < 𝑥 < ∞

𝑑
2. 𝐹′ 𝑥 = 𝐹 𝑥 =𝑓 𝑥 ≥0
𝑑𝑥

⇒ 𝐹(𝑥) is non-decreasing function of x.

𝑥 −∞
3. 𝐹 −∞ = lim 𝐹 𝑥 = lim ‫׬‬−∞ 𝑓 𝑥 𝑑𝑥 = ‫׬‬−∞ 𝑓 𝑥 𝑑𝑥 = 0
𝑥→−∞ 𝑥→−∞

𝑥 ∞
𝐹 +∞ = lim 𝐹 𝑥 = lim ‫׬‬−∞ 𝑓 𝑥 𝑑𝑥 = ‫׬‬−∞ 𝑓 𝑥 𝑑𝑥 = 1
𝑥→∞ 𝑥→∞

4. 𝐹(𝑥) is continuous function of 𝑥 on the right.

5. The discontinuities of 𝐹(𝑥) are at the most countable.

The diameter of an electric cable, say 𝑋, is assumed to be a continuous r.v. with
p.d.f :𝑓 𝑥 = 6𝑥(1 − 𝑥); 0 ≤ 𝑥 ≤ 1

(i) Check that 𝑓(𝑥) is 𝑝. 𝑑. 𝑓,

(ii) Determine a number 𝑏 such that 𝑃 𝑋 < 𝑏 = 𝑃(𝑋 > 𝑏).

1 1
i)  f ( x ) dx = 6  x (1 − x ) dx
0 0

1 2 3 1

= 6  ( x − x ) dx = 6
x x
2
− =1
0
2 3 0

 f ( x ) is the p.d . f of r.v. X

P ( X  b) = P ( X  b)
b 1

 f ( x ) dx =  f ( x ) dx
0 b
b 1
6  x (1 − x ) dx = 6  x (1 − x ) dx
0 b
b 1
x x 
2
x
3
x  2 3

 2 − 3  = 2 − 3 
 0  b
4b3 − 6b 2 + 1 = 0
 1
− ( − 4b − 2 ) = 0
2
 b 4b
 2
1
b= , b 1.366, -0.3660
2
A continuous random variable X has a p.d.f. 𝑓 𝑥 =
2
3𝑥 , 0 ≤ 𝑥 ≤ 1. Find a and b such that
i. 𝑃 𝑋 ≤ 𝑎 = 𝑃 𝑋 > 𝑎
ii. 𝑃 𝑋 > 𝑏 = 0.05 1
1
i)Since P ( X  a ) = P ( X  a ) ii) P ( X  b ) = 0.05   f ( x ) dx =
b
20
1
 P( X  a) = P( X  a) = 1
2 1  19  3
a
1 1− b =
3
b= 
P ( X  a ) =  f ( x ) dx = 20  20 
0
2
1
1 1 3
a = a= 
3

2 2
Let 𝑋 be a continuous random variable with p.d.f. :

𝑎𝑥 ; 0≤𝑥≤1
𝑎 ; 1≤𝑥≤2
𝑓 𝑥 = ൞
−𝑎𝑥 + 3𝑎 ; 2 ≤ 𝑥 ≤ 3
0 ; 𝑒𝑙𝑠𝑒𝑤ℎ𝑒𝑟𝑒
(i) Determine the constant 𝒂.
(ii) Compute 𝑃(𝑋 ≤ 1.5).

i)  f ( x ) dx = 1
−
0 1 2 3 

 f ( x ) dx +  f ( x ) dx +  f ( x ) dx +  f ( x ) dx +  f ( x ) dx = 1
− 0 1 2 3

1
a=
2
 1.5 0 1 1.5
ii) P ( X  1.5 ) =  f ( x ) dx =  f ( x ) dx +  f ( x ) dx +  f ( x ) dx
− − 0 1
1 1.5
=  ax dx +  a dx
0 1

3
a=
4