Probability

Distribution
Contents 2

▶ Probability Distribution
▶ Discrete and Continuous Distribution
▶ Mathematical Expectations
▶ Binomial Distribution
▶ Poisson Distribution
▶ Uniform Distribution
▶ Exponential Distribution
▶ Normal Distribution, Normal Curve and Its Characteristics
▶ Area Under The Normal Curve
Random Variable 3

A random variable is a variable that takes on numerical values as a result of a

random experiment or measurement; associates a numerical value with each
possible outcome.

The differences between variable and random variable are-

• Random variable always takes numerical values
• There is a probability associated with each possible values

Random variable is denoted by capital letters such as X, Y, Z etc.

And the possible outcomes are denoted by small letters such as x, y, z etc.
Random Variable 4

Example 1:
A coin is tossed. It has two possible outcomes- Head and Tail.

𝐻, 𝑖𝑓𝐻𝑒𝑎𝑑 𝑎𝑝𝑝𝑒𝑎𝑟𝑠
Consider a variable, X= outcome of a coin toss= ቊ𝑇,
𝑖𝑓 𝑇𝑎𝑖𝑙 𝑎𝑝𝑝𝑒𝑎𝑟𝑠
Here, S= {H, T}.
But, these are not numerical values.
Random Variable 5

Example 1(contd.):
Consider a variable, X= Number of heads obtained in a trial
1, 𝑖𝑓 𝐻𝑒𝑎𝑑 𝑎𝑝𝑝𝑒𝑎𝑟𝑠
Then, 𝑋 = ቊ
0, 𝑖𝑓 𝑇𝑎𝑖𝑙 𝑎𝑝𝑝𝑒𝑎𝑟𝑠
For a fair coin, we can write, P(X=1) = ½ and P(X=0) = ½
So, X is a random variable.
Random Variable 6

Types of random variable:

Random Variable

Discrete Random Variable Continuous Random Variable

A random variable defined over A random variable defined over a

a discrete sample space continuous sample space
Random Variable 7

Examples:
Discrete Random Variable:
1. X= Number of correct answers in a 100-MCQ test= 0, 1, 2, …, 100
2. X= Number of cars passing a toll both in a day= 0, 1, 2, …, ∞
3. X= Number of balls required to take the first wicket = 1, 2, 3, …, ∞

Continuous Random Variable:

1. X= Weight of a person. 0<X<∞
2. X= Monthly Profit. -∞<X<∞
Probability Distributions 8

Distribution of the probabilities among the different values of a random variable.

Discrete probability distribution- probability distribution of a discrete random

variable

Continuous probability distribution- probability distribution of a continuous

random variable
Probability Distributions 9

Examples:
Discrete probability distribution-
• Tossing a coin 2 times.
X= Number of Heads appeared
S= {HH, HT, TH, TT}

x 0 1 2
P(x) ¼ 2/4 ¼
Probability Distributions 10

Different types of probability distributions:

Discrete probability distribution-

Probability
1. Bernoulli Distribution
2. Binomial Distribution
3. Poisson Distribution etc.
Continuous probability distribution- Values

1. Uniform Distribution

Probability
2. Normal Distribution
3. Exponential Distribution
4. t-distribution etc.
Values
PMF and PDF 11

Probability Mass Function (pmf)- the probability distribution function of a

discrete random variable X is called a pmf and is denoted by p(x)

Probability Density Function (pdf)- the probability distribution function of a

continuous random variable X is called a pdf and is denoted by f(x)
Mathematical Expectations 12

• For a discrete random variable X with pmf p(x), the mathematical expectation
of X is-
𝜇=𝐸 𝑋 =෍ 𝑥𝑝 𝑥
𝑥
• For a continuous random variable X with pdf f(x), the mathematical
expectation of X is-

𝜇 = 𝐸 𝑋 = න𝑥 𝑓 𝑥
𝑥

Mathematical expectation is also known as population mean or expected value.

Mathematical Expectations 13

෍ 𝑥2 𝑝 𝑥 , 𝑖𝑓 𝑥 𝑖𝑠 𝑎 𝑑𝑖𝑠𝑐𝑟𝑒𝑡𝑒 𝑟. 𝑣.
𝑥
𝐸 𝑋2 =
න𝑥 2 𝑓 𝑥 , 𝑖𝑓 𝑥 𝑖𝑠 𝑎 𝑐𝑜𝑛𝑡𝑖𝑛𝑢𝑜𝑢𝑠 𝑟. 𝑣.
𝑥

Variance:
𝜎 2 = 𝑉𝑎𝑟 𝑋 = 𝐸 𝑋 − 𝐸 𝑋 2 = 𝐸 𝑋2 − 𝐸 𝑋 2 = 𝐸 𝑋 2 − 𝜇2

Standard deviation: 𝜎 = 𝑉𝑎𝑟 𝑋

Properties of Mathematical Expectations 14

Let, c is a constant number

X and Y are two independent random variables

1. E(c) = c 1. Var(c) = 0
2. E(c X) = c E(x) 2. Var(c X) = c2 Var(x)
3. E(X + c) = E(x) + c 3. Var(X + c) = Var(x)
4. E(X+Y) = E(X) + E(Y) 4. Var(X+Y) = Var(X) + Var(Y)
5. E(X-Y) = E(X) - E(Y) 5. Var(X-Y) = Var(X) + Var(Y)
6. E(XY) = E(X) . E(Y)
Mathematical Expectation 15

Example 2-
A company estimates the net profit on a new product, it is launching, to be Rs. 3
million during first year, if it is ‘successful’, Rs. 1 million if it is ‘moderately
successful’, and a loss of Rs. 1 million if it is ‘unsuccessful’.
The company assigns the following probabilities to first year prospects for the
product-
Successful: 0.25, Moderately successful: 0.40, and Unsuccessful: 0.35

What are the expected value and standard deviation of the first year net profit
for the product? Also, find the expected value of net profit if there is a fixed cost
of Rs. 0.2 million, whatever the success status is.
Mathematical Expectation 16

Solution-
Let,
X= Net profit on the new product in the 1st year (Rs. Million)
Given that,
x 3 1 -1
P(x) 0.25 0.4 0.35

Expected net profit, 𝐸 𝑋 = ∑𝑥 𝑝 𝑥 = 3 ∗ 0.25 + 1 ∗ 0.4 + −1 ∗ 0.35

= 0.8 𝑚𝑖𝑙𝑙𝑖𝑜𝑛
Mathematical Expectation 17

Solution (contd.)-

𝐸 𝑋 2 = ∑ 𝑥 2 𝑝 𝑥 = 32 ∗ 0.25 + 12 ∗ 0.4 + −1 2 ∗ 0.35

= 9 ∗ 0.25 + 1 ∗ 0.4 + 1 ∗ 0.35 = 3

𝑉𝑎𝑟 𝑋 = 𝐸 𝑋 2 − 𝐸 𝑋 2 = 3 − 0.82 = 2.36

∴ 𝑆𝐷 𝑋 = 𝑉𝑎𝑟 𝑋 = 2.36 = 1.54 𝑚𝑖𝑙𝑙𝑖𝑜𝑛

If there is a fixed cost of Rs. 0.2 million, then expected net profit-
𝐸 𝑋 − 0.2 = 𝐸 𝑋 − 0.2 = 0.8 − 0.2 = 0.6 𝑚𝑖𝑙𝑙𝑖𝑜𝑛
Binomial Distribution 18

Bernoulli trial:
A trial that has only two possible outcomes (often called ‘Success’ and ‘Failure’)

Outcome Success failure

Probability p 1-p

Let,
• n independent Bernoulli trials are performed
• Each trial has the same probability of success, p
Binomial Distribution 19

Let,
X= number of success in n trials

Then, X is a binomial random variable with distribution function (pmf),

𝑝 𝑥 = 𝑛𝐶𝑥 𝑝 𝑥 1 − 𝑝 𝑛 − ; 𝑥 = 0, 1, 2, … , 𝑛
𝑛!
= 𝑝𝑥 1 − 𝑝 𝑛 −
𝑛 − 𝑥 ! 𝑥!

Here, n!= n(n-1)(n-2) …1 0!=1 1!=1 2!= 2*1= 2 3!= 3*2*1=6

We write it as, 𝑋 ~ 𝑏𝑖𝑛𝑜𝑚𝑖𝑎𝑙 𝑛, 𝑝

Binomial Distribution 20

Mean of the binomial distribution, 𝜇 = 𝐸 𝑋 = ∑𝑥 𝑝 𝑥 = 𝑛𝑝

Variance of binomial distribution, 𝜎 2 = 𝐸 𝑋 2 − 𝜇 2 = 𝑛𝑝 1 − 𝑝 = 𝑛𝑝𝑞

Standard deviation of binomial distribution, 𝜎 = 𝑛𝑝𝑞

Binomial Distribution 21

Example 3:
There are 3 multiple choice questions in a MCQ test. Each MCQ consists of four
possible choices and only one of them is correct. If an examinee answers those
MCQ randomly (without knowing the correct answers)
a. What is the probability that exactly any two of the answers will be correct?
b. What is the probability that at least two of the answers will be correct?
c. What is the probability that at most two of the answers will be correct?
d. What will be the average or expected number of correct answers?
e. Also, find the standard deviation of number of correct answers.
Binomial Distribution 22

Solution:
Let,
X= number of correct answers selected in 3 MCQs

Here, p = probability of selecting correct answer per question = ¼ = 0.25

∴ 𝑋 ~ 𝑏𝑖𝑛𝑜𝑚𝑖𝑎𝑙 𝑛 = 2, 𝑝 = 0.25

𝑝 𝑥 = 3𝐶𝑥 0.25 𝑥 1 − 0.25 3−𝑥 ; 𝑥 = 0, 1, 2, 3

3!
= 0.25 𝑥 0.75
3 − 𝑥 ! 𝑥! 𝑛−
Binomial Distribution 23

Solution (contd.):
a. probability that exactly any two of the answers will be correct-
3!
𝑃 𝑋=2 = 0.25 2 0.75 3−2
3 − 2 ! 2!
3! 2 1
3∗2∗1
= 0.25 0.75 = ∗ 0.0625 ∗ 0.75 = 0.141
1! 2! 1∗ 2∗1

b. probability that at least two of the answers will be correct-

𝑃 𝑋 ≥ 2 =𝑃 𝑋 =2 +𝑃 𝑋 =3
3! 2 3−2 +
3!
= 0.25 0.75 0.25 3 0.75 3−3
3 − 2 ! 2! 3 − 3 ! 3!
3! 3!
= 0.25 2 0.75 1 + 0.25 3 0.75 0 = 0.141 + 0.016 = 0.157
1! 2! 0! 3!
Binomial Distribution 24

Solution (contd.):
c. probability that at most two of the answers will be correct-
𝑃 𝑋 ≤ 2 =𝑃 𝑋 =0 +𝑃 𝑋 = 1 +𝑃 𝑋 =2
3! 3!
= 0.25 0 0.75 3−0 + 0.25 1 0.75 3−1
3 − 0 ! 0! 3 − 1 ! 1!
3!
+ 0.25 2 0.75 3−2
3 − 2 ! 2!
3! 3! 3!
= 0.25 0 0.75 3 + 0.25 1 0.75 2 + 0.25 2 0.75 1
3! 0! 2! 1! 1! 2!
= 0.422 + 0.422 + 0.141 = 0.985
d. 𝐸 𝑋 = 𝑛𝑝 = 3 ∗ .25 = 0.75
e. 𝑆𝐷 𝑋 = 𝑛𝑝𝑞 = 3 ∗ 0.25 ∗ 0.75 = 0.75
Binomial Distribution 25

X 0 1 2 3
P(x) 0.422 0.422 0.141 0.016

Probability 0.4

0.2

0
0 1 2 3
Number of correct answers
Poisson Distribution 26

Let,
X= a random variable usually counts or number of occurrences

Then, X is a Poisson random variable with distribution function (pmf),

𝑒 −𝜆 𝜆𝑥
𝑝 𝑥 = ; 𝑥 = 0, 1, 2, …
𝑥!

We write it as, 𝑋 ~ 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 𝜆

Poisson Distribution 27

Mean of the Poisson distribution, 𝜇 = 𝐸 𝑋 = ∑𝑥 𝑝 𝑥 = 𝜆

Variance of Poisson distribution, 𝜎 2 = 𝐸 𝑋 2 − 𝜇2 = 𝜆

Standard deviation of Poisson distribution, 𝜎 = 𝜆

Poisson Distribution 28

Example 4:
The average number of errors on a page of a certain magazine is 0.2.
What is the probability that the next page (or a randomly selected page)
you read contains
i. 0 (zero) error?
ii. 2 or more errors?
iii. What is the average error per page?
iv. Also, find standard deviation of the number of errors.
Poisson Distribution 29

Solution:
Let,
X= number of errors in a page

Here, λ = average number of errors per page= 0.2

∴ 𝑋 ~ 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 𝜆 = 0.2

𝑒 −𝜆 𝜆𝑥
𝑝 𝑥 = 𝑥! ; 𝑥 = 0, 1, 2, …
𝑒−0.2 0.2𝑥
=
𝑥!
Poisson Distribution 30

Solution:
e−0.2 0.2 x e−0.2 0.20 e −0.2 ∗1
i. P X=0 = = = = 0.8187
X! 0! 1

ii. P X≥2 =1−P X<2 =1− P X=0 +P X=1

e−0.2 0.20 e−0.2 0.21
= + = 1 − e−0.2 + e−0.2 ∗ 0.2 = 0.01756
0! 1!
iii. Average number of errors, 𝐸 𝑋 = 𝜆 = 0.2

iv. Standard deviation, 𝑆𝐷 𝑋 = 𝜆 = 0.2 = 0.45

Poisson Distribution 31

x p(x)
0 0.82
1 0.16 Poisson (λ=0.2)
2 0.02 0.90
3 0.00 0.80
4 0.00
5 0.00
0.70
6 0.00 0.60
7 0.00 0.50
8 0.00 0.40
9 0.00
0.30
10 0.00
15 0.00 0.20
20 0.00 0.10
30 0.00 0.00
40 0.00
0 1 2 3 4 5 6 7 8 9 10 15 20 30 40 50 100
50 0.00
100 0.00 p(x)
Uniform Distribution 32

Let,
X is a continuous random variable

Then, if X has a probability density function (pdf),

1
𝑓 𝑥 = ; 𝑎<𝑥<𝑏
𝑏−𝑎 1
𝑏−𝑎
We write it as, 𝑋 ~ 𝑢𝑛𝑖𝑓𝑜𝑟𝑚 𝑎, 𝑏
Mean, 𝐸 𝑋 = 𝑎+𝑏
2
𝑏−𝑎 2 a b
Variance, 𝑉 𝑋 =
12
x
Uniform Distribution 33

Example 5:
The waiting time (in minutes) for train is uniform (10, 50).
Find-
a. The probability that you have to wait at least 20 minutes.
b. Average waiting time.
c. Standard deviation of waiting time.
Uniform Distribution 34

Solution:
Let,
X= waiting time (in minutes)

∴ 𝑋 ~ 𝑢𝑛𝑖𝑓𝑜𝑟𝑚 10, 50

1
𝑓 𝑥 = 𝑏−𝑎 ; 𝑎 <𝑥 <𝑏
1
= 50 − 10 ; 10 < 𝑥 < 50
1
=
40
Uniform Distribution 35

Solution:
a. Probability that you have to wait at least 20 minutes-
50 50
1 1 50
P 𝑋 ≥ 20 = න 𝑓 𝑥 𝑑𝑥 = න 𝑑𝑥 = න 1 𝑑𝑥
20 20 40 40 20
1 1 30 3
= 𝑥 50
20 = 50 − 20 = = = 0.75
40 40 40 4

b. Average waiting time-

𝑎 + 𝑏 10 + 50 60
𝐸 𝑋 = = = = 30 𝑚𝑖𝑛𝑢𝑡𝑒𝑠
2 2 2
Uniform Distribution 36

Solution:
c. Standard deviation of waiting time-
𝑏−𝑎 2 50 − 10 2 40 2
𝑆𝐷 𝑋 = = = = 3.65 𝑚𝑖𝑛𝑢𝑡𝑒𝑠
12 12 12
Exponential Distribution 37

A continuous random variable X is said to follow

exponential distribution, if its pdf is,

𝑓 𝑥 = 𝜆𝑒−𝜆𝑥; 𝑥 > 0, 𝑤ℎ𝑒𝑟𝑒 𝜆 > 0

We write as, 𝑋~ exp 𝜆

Here, X is usually time until certain event occurs.

1
Mean, 𝐸 𝑋 =
𝜆

Variance, 𝑉 𝑋 = 1
𝜆2
Exponential Distribution 38

Example 6:
Average time required to repair a machine is 0.5 hours. What is the
probability that the next repair will take more than 2 hours?
Solution:

Let, X= time required to repair the machine

1 1
∴ 𝑋~ exp 𝜆 = 2 𝑠𝑖𝑛𝑐𝑒, 𝐸 𝑋 = = 0.5 ⇒ 𝜆 = =2
𝜆 0.5
∞ ∞
Pr 𝑋 > 2 = න 𝜆𝑒−𝜆𝑥𝑑𝑥 = −𝑒−𝜆𝑥 2 = −0 + 𝑒−2𝜆 = 𝑒−2𝜆 = 𝑒−2×2
2
= 𝑒−4 = 0.0183
Normal Distribution 39

Let,
X is a continuous random variable

Then, if X has a probability density function (pdf),

1 1
− 2 𝑥−𝜇 2
𝑓 𝑥 = 𝑒 2𝜎 ; −∞ < 𝑥 < ∞
𝜎 2𝜋

We write it as, 𝑋 ~ 𝑁 𝜇, 𝜎 2
Mean, 𝐸 𝑋 = 𝜇
-∞ μ ∞
Variance, 𝑉 𝑋 = 𝜎2
x
Standard Normal Distribution 40

Let,
𝑋−𝜇
𝑍=
𝜎
Then, Mean, 𝐸 𝑍 = 0
Variance, 𝑉 𝑍 = 1
And, if Z has a probability density function (pdf),

1 1
−2 𝑧 2
𝑓 𝑧 = 𝑒 ; −∞ < 𝑧 < ∞
2𝜋

-∞ 0 ∞
We write it as, 𝑍 ~ 𝑁 0, 1
z
Characteristics of a Normal Distribution 41

1. Mean= Median = Mode

2. Symmetric and Mesokurtic
3. Bell-shaped curve
4. The area under the curve lying between μ±σ is 68.27%
of the total area
5. The area under the curve lying between μ±2σ is
95.45% of the total area
6. The area under the curve lying between μ±3σ is
99.73% of the total area

-∞ μ-σ μ μ+σ ∞
Characteristics of a Normal Distribution 42

50% 50% 50% 50%

-∞ μ ∞ -∞ 0 ∞
x z

P[X<μ] = P[X> μ] = 0.5 P[Z<0] = P[Z> 0] = 0.5

P[X<-x] = P[X>x] P[Z<-z] = P[Z>z]

Normal Distribution Table 43
Z-table
• Normal distribution table provides probabilities for N(0,1) i.e. for standard
normal distribution
• Usually, normal table gives P[0 < Z < z] for positive values of Z.
• For other values, we can use the property of symmetry with median 0 of
standard normal distribution
• To find probabilities for a normal random variable X, we can transform the
probability statement about X in terms of probability statement for Z and then
calculate the probability using the standard normal distribution table or Z-
table
𝑋−𝜇 𝑎−𝜇 𝑎−𝜇
𝑃 𝑋<𝑎 =𝑃 < =𝑃 𝑍<
𝜎 𝜎 𝜎
Finding Area Under the Normal Curve 44
using Z-table
Example 7:
The number of viewers of a TV show per week has a mean of 29 million with a
standard deviation of 5 million. Assume that, the number of viewers of that show
follows a normal distribution.
What is the probability that, next week’s show will-
a. Have between 30 and 34 million viewers?
b. Have at least 23 million viewers?
c. Exceed 40 million viewers?
Finding Area Under the Normal Curve 45
using Z-table
Solution:
Let, X= Number of viewers of the show per week (in million)
-∞ 0 0.2 1 ∞
∴ 𝑋 ~ 𝑁 𝜇, 𝜎 2 z

a. the probability that, next week’s show will have between 30 and 34 million viewers-

30 − 𝜇 𝑋 − 𝜇 34 − 𝜇 30 − 29 𝑋 − 𝜇 34 − 29
𝑃 30 ≤ 𝑋 ≤ 34 = 𝑃 ≤ ≤ =𝑃 ≤ ≤
𝜎 𝜎 𝜎 5 𝜎 5
= 𝑃 0.20 ≤ 𝑍 ≤ 1 = 𝑃 0 ≤ 𝑍 ≤ 1 − 𝑃 0 ≤ 𝑍 ≤ 0.2 = 0.3413 − 0.0793
= 0.262
Finding Area Under the Normal Curve 46
using Z-table
Solution (contd.):
b. the probability that, next week’s show will have at least 23 million
viewers-
𝑋 − 𝜇 23 − 𝜇 𝑋 − 𝜇 23 − 29
𝑃 𝑋 ≥ 23 = 𝑃 ≥ =𝑃 ≥
𝜎 𝜎 𝜎 5 -∞ -1.2 0 ∞
= 𝑃 𝑍 ≥ −1.2 = 𝑃 −1.2 ≤ 𝑍 ≤ 0 + 𝑃 𝑍 ≥ 0 = 0.3849 + 0.5 z
= 0.8849
c. the probability that, next week’s show will exceed 40 million
viewers-

𝑋 − 𝜇 40 − 𝜇 𝑋 − 𝜇 40 − 29
𝑃 𝑋 > 40 = 𝑃 > =𝑃 >
𝜎 𝜎 𝜎 5 0 2.2 ∞
= 𝑃 𝑍 > 2.2 = 𝑃 𝑍 ≥ 0 − 𝑃 0 ≤ 𝑍 ≤ 2.2 = 0.5 − 0.4861 = 0.0139 -∞ z
Finding Area Under the Normal Curve 47
using Z-table
Example 8:
a. For what value of ‘a’, P[Z≤a] = 0.95?
b. For what value of ‘a’, P[Z≥a] = 0.05?
c. For what value of ‘a’, P[Z≤a] = 0.975?

Solution:
a. P[Z≤a] = 0.95
Or, P[Z≤0] + P[0<Z≤a] = 0.95
0.95
Or, 0.5+ P[0<Z≤a] =0.95
Or, P[0<Z≤a] =0.95-0.5= 0.45
-∞ 0 1.645 ∞
For a= 1.645, P[0<Z≤a] =0.45 z
Finding Area Under the Normal Curve 48
using Z-table
Solution (contd.): Solution (contd.):
b. P[Z≥a] = 0.05 c. P[Z≤a] = 0.975
Or, P[Z ≥ 0] - P[0<Z≤a] = 0.05 Or, P[Z≤0] + P[0<Z≤a] = 0.975
Or, 0.5- P[0<Z≤a] =0.05 Or, 0.5+ P[0<Z≤a] =0.975
Or, P[0<Z≤a] =0.5-0.05= 0.45 Or, P[0<Z≤a] =0.975-0.5= 0.475
For a= 1.645, P[0<Z≤a] =0.45 For a= 1.96, P[0<Z≤a] =0.475

0.975

0.05
-∞ 0 1.645 ∞ -∞ 0 1.96 ∞
z z