Probability

Contents 2

 Probability
 Basic Probability Concepts
 Mathematical Properties of Probability
 Rules of Probability
 Bayes’ Theorem
 Finding Probability Using Contingency Table
Probability 3

Probability is the likeliness of occurring any event(s).

Some related terms 4

Deterministic experiment Vs Random experiment:

An experiment whose outcome is predictable in advance is called
deterministic experiment. Everyone conducting that experiment
will get the same outcome.

An experiment whose outcome is not predictable with certainty in

advance is called a random experiment. If a random experiment is
performed then one of many possible outcomes will occur.
Some related terms 5

Deterministic experiment Vs Random experiment:

Example:
Deterministic experiment:
 Measuring linear distance from Dhaka to Chittagong.
 Measuring length of a scale.

Random experiment:
 Measuring weight of a person at different times.
 Tossing a coin.
Some related terms 6

Sample Space:
The set of all possible outcomes of a random experiment is called
sample space of that experiment and is denoted by S. Each
individual outcome is called a sample point.

For example;
throwing a dice- S= {1, 2, 3, 4, 5, 6 }
Lifetime of a lightbulb- S= {x|0≤x<∞} = [0,∞)
Some related terms 7

Event:
Any subset E of a sample space S is an event.

For example;
Dice throw experiment- S= {1, 2, 3, 4, 5, 6 }
Say number 2 turned up in a throw. Then we will say, event E= {2}
has occurred.
Some related terms 8

Mutually Exclusive events:

Two events are called mutually exclusive if both the events cannot
occur simultaneously in a single trial. In other words, if one of those
events occurs, the other event will not occur.

For example;
in a trial of coin toss experiment, event E1= {Head} and event
E2={tail} will not occur simultaneously. So, E1 and E2 are mutually
exclusive events.
On a day, Event E1={Rain} & event E2={Sunny} may occur
simultaneously. These are not mutually exclusive events.
Some related terms 9

Collectively exhaustive events:

Collectively exhaustive events are those, which includes all possible
outcomes.

For example;
In a coin tossing experiment events E1= {Head} and event E2={tail}
are collectively exhaustive, because together they comprise the all
the outcomes that are possible in a coin tossing experiment. There
are no other possible outcomes of this experiment than these two.
Some related terms 10

Equally likely events:

The events of a random experiment are called equally likely if the
chance of occurring those events are all equal.

For example;
In a coin tossing experiment, the events E1= {Head} and event
E2={tail} are equally likely, because the chance of occurring E1 is as
same as occurring E2.
On a day, Event E1={Rain} & event E2={No rain} may not be equally
likely.
Some related terms 11

Disjoint events
Disjoint events:
Two events are called disjoint, if they E1
E2
have no common elements between
them.

Mutually exclusive events are disjoint

events.
E1 E2

Joint events
Approaches of assigning probability 12

 At first we identify the sample space S of the random

experiment.
 We then define our favorable event and assign probability to the
event using one of the following 3 basic approaches-
 Classical approach
 Frequency approach
 Subjective approach
Approaches of assigning probability 13

Classical approach:
(when the outcomes are equally likely, mutually exclusive and
collectively exhaustive)
 If the sample space of a random experiment has a finite number
(ns) of outcomes
 nE of these outcomes are favorable to an event E

Then, the probability of occurring event E, denoted by P(E) is-

𝑛𝐸
𝑃 𝐸 =
𝑛𝑆
Approaches of assigning probability 14

Classical approach:
For example;
Dice throwing experiment-
S= {1, 2, 3, 4, 5, 6}
Consider two events, E1= {2} and E2= {2, 4, 6}
Here, nE1= 1 and nE2= 3. Also, nS= 6
𝑛𝐸1 1
Therefore, probability of occurring event E1 is, 𝑃 𝐸1 = =
𝑛𝑆 6
𝑛𝐸2 3 1
probability of occurring event E2 is, 𝑃 𝐸2 = = =
𝑛𝑆 6 2
Approaches of assigning probability 15

Frequency approach:
If an experiment is repeated n times under the same conditions and
event E occurs f times out of n times, then
𝑓
𝑃 𝐸 = lim
𝑛→∝ 𝑛

That is, when n is very large, P(E) is very close to the relative
frequency of event E.
Approaches of assigning probability 16

Frequency approach:
For example;
In a dice throwing experiment- S= {1, 2, 3, 4, 5, 6}
And our favorable event is E= {2}

Let, 2 occurred a total of 998 times out of total 6000 trials.

998 1
Therefore 𝑃 𝐸 = lim ≈
𝑛→∝ 6000 6
Approaches of assigning probability 17

Subjective approach:
Based on the judgement (personal experience, prior information
and belief etc.), one can assign probability to an event E of a
random experiment.

For example; on a day of summer someone made a statement on

probability that rain will occur on that day is .70, based on his
previous experience.
Axioms of probability 18

Valid probabilities will follow 3 axioms-

Axiom 1: (Axiom of positivizes) : 0≤ P(E) ≤1
Axiom 2: (Axiom of certainty) : P(S) = 1
Axiom 3: (Axiom of additivity) : For a sequence of disjoint events E1,
E2, …, En-
𝑛 𝑛

𝑃 𝐸𝑖 = 𝑃 𝐸𝑖
𝑖=1 𝑖=1
Example 1 19

In a community of 400 people, 20 people has a particular disease. If

a person is selected randomly from that community, what is the
probability that he/ she does not has the disease?
Example 1 20

In a community of 400 people, 20 people has a particular disease. If

a person is selected randomly from that community, what is the
probability that he/ she does not has the disease?

Solution:
Let, D= the randomly selected person has the disease
𝑓 20
ℎ𝑒𝑟𝑒, 𝑃 𝐷 = = = .05
𝑛 400
∴ 𝑃 𝐷𝑐 = 1 − .05 = 0.95
So, the probability that he/ she does not have the disease is 0.95
Probability Laws: 21
Addition Law
• For disjoint events A and B-
The probability that, either event A or event B will occur is,
𝑃 𝐴∪𝐵 =𝑃 𝐴 +𝑃 𝐵
• For disjoint events A, B, C, … , and Z-
The probability that, either event A or event B or event C or … or
event Z will occur is,
𝑃 𝐴 ∪ 𝐵 ∪ 𝐶 ∪ ⋯∪ 𝑍 = 𝑃 𝐴 + 𝑃 𝐵 + 𝑃 𝐶 + …+ 𝑃 𝑍
Probability Laws: 22
Addition Law
• For joint events A and B-
The probability that, either event A or event B or both will occur is,
𝑃 𝐴∪𝐵 =𝑃 𝐴 +𝑃 𝐵 −𝑃 𝐴∩𝐵
• For joint events A, B, and C
The probability that, either event A or event B or event C or any two
of them or all will occur is,
𝑃 𝐴∪𝐵∪𝐶
=𝑃 𝐴 +𝑃 𝐵 +𝑃 𝐶 −𝑃 𝐴∩𝐵 −𝑃 𝐵∩𝐶 −𝑃 𝐴∩𝐶
+𝑃 𝐴∩𝐵∩𝐶
Probability Laws: 23
Addition Law
Example 2:
In a company, 60% of the employees have motorcycle, 40% has
private car and 20% has both.
If an employee is selected randomly from that company, then
a. What is the probability that the employee has either motorcycle
or private car?
b. What is the probability that the employee has neither
motorcycle nor private car?
Probability Laws: 24
Addition Law
Solution:
Let,
M= the randomly selected employee has motorcycle C M
C= the randomly selected employee has car

60 40
Here, 𝑃 𝑀 = = 0.6, 𝑃 𝐶 = = 0.4
100 100
20
𝑃 𝑀∩𝐶 = = 0.2
100
Probability Laws: 25
Addition Law
Solution(contd.):
C M
a. probability that the person has either
motorcycle or private car is-
𝑃 𝑀∪𝐶 =𝑃 𝑀 +𝑃 𝐶 −𝑃 𝑀∩𝐶
= 0.6 + 0.4 − 0.2 = 0.8
Probability Laws: 26
Addition Law
Solution(contd.):
C M
a. probability that the person has either
motorcycle or private car is-
𝑃 𝑀∪𝐶 =𝑃 𝑀 +𝑃 𝐶 −𝑃 𝑀∩𝐶
= 0.6 + 0.4 − 0.2 = 0.8
b. probability that the person has neither
motorcycle nor private car is- C M

𝑃 𝑀∪𝐶 𝑐 =1−𝑃 𝑀∪𝐶

= 1 − 0.8 = 0.2
Probability Laws: 27

Conditional Probability:
The conditional probability of an event E, given that another event F
has already happened is,
𝑃 𝐸∩𝐹
𝑃 𝐸|𝐹 = 𝑓𝑜𝑟 𝑃 𝐹 > 0
𝑃 𝐹
Probability Laws: 28

Conditional Probability:
Illustration: A dice throwing experiment.
S= {1, 2, 3, 4, 5, 6}
P(x) Probability
P(1)
P(2)
P(3)
P(4)
P(5)
P(6)
Total
Probability Laws: 29

Conditional Probability:
Illustration: A dice throwing experiment.
S= {1, 2, 3, 4, 5, 6}
P(x) Probability
P(1) 1/6
P(2) 1/6
P(3) 1/6
P(4) 1/6
P(5) 1/6
P(6) 1/6
Total 1
Probability Laws: 30

Conditional Probability:
Illustration(contd.):
Now assume that, someone said that, in that throw an odd number
appeared on the dice.

That is, event F= {1, 3, 5} has happened.

So, our sample space S has reduced to S*= {1, 3, 5} = F

Probability Laws: 31

Conditional Probability:
Illustration(contd.):
P(x|F) Probability
Now, S*= {1, 3, 5} P(1|F)
P(2|F)
P(3|F)
Since, these outcomes were
P(4|F)
equally likely in the original
P(5|F)
sample space, they are equally
P(6|F)
likely in revised sample space
Total
Probability Laws: 32

Conditional Probability:
Illustration(contd.):
P(x|F) Probability
Now, S*= {1, 3, 5} P(1|F) 1/3
P(2|F) 0
P(3|F) 1/3
Since, these outcomes were
P(4|F) 0
equally likely in the original
P(5|F) 1/3
sample space, they are equally
P(6|F) 0
likely in revised sample space
Total 1
 Probability Laws: 33

Conditional Probability:
Illustration(contd.):
Now, S*= {1, 3, 5}
F
E

Let, E= {1} and F= {1, 3, 5}

𝑃 𝐸∩𝐹
𝑃 𝐸|𝐹 =
𝑃 𝐹
Probability Laws: 34

Conditional Probability:
Illustration(contd.):
Now, S*= {1, 3, 5} F
E

Let, E= {1} and F= {1, 3, 5}

1
𝑃 𝐸∩𝐹 6 1
𝑃 𝐸|𝐹 = = =
𝑃 𝐹 3 3
6
Probability Laws: 35

Conditional Probability:
Illustration(contd.):
Now, S*= {1, 3, 5}
F
E
Let, E= {2} and F= {1, 3, 5}

𝑃 𝐸∩𝐹 0
𝑃 𝐸|𝐹 = = =0
𝑃 𝐹 3
6
Probability Laws: 36

Example 3:
In a company, 60% of the employees have motorcycle, 40% has
private car and 20% has both.
If an employee is selected randomly from that company, then
a. What is the probability that the employee has a car?
b. If it is known that the employee has a motorcycle, then what is
the probability that the employee also has a car?
Probability Laws: 37

Solution:
Let,
M= the randomly selected employee has C M
motorcycle
C= the randomly selected employee has car

60 40
Here, 𝑃 𝑀 = = 0.6, 𝑃 𝐶 = = 0.4
100 100
20
𝑃 𝑀∩𝐶 = = 0.2
100
Probability Laws: 38

Solution(contd.):
a. probability that the employee has a car is- C M
𝑃 𝐶 = 0.4
Probability Laws: 39

Solution(contd.):
a. probability that the employee has a car is- C M
𝑃 𝐶 = 0.4

b. If it is known that the employee has a

motorcycle, then what is the probability C M
that the employee also has a car is,
𝑃 𝑀∩𝐶 0.2 1
𝑃 𝐶|𝑀 = = = = 0.33
𝑃 𝑀 0.6 3
Probability Laws: 40
Multiplication Law
• For two dependent events E and F-
The probability that, both event E and event F will occur
simultaneously is,
𝑃 𝐸 ∩ 𝐹 = 𝑃 𝐸|𝐹 𝑃 𝐹
Here, occurrence of event E depends on occurrence of event F.

• For two independent events E and F-

The probability that, both event E and event F will occur
simultaneously is,
𝑃 𝐸∩𝐹 =𝑃 𝐸 𝑃 𝐹
Probability Laws: 41
Multiplication Law
Example 4:
In rainy season, it rains 70% of the days in Bangladesh. When it rains,
80% times it makes thunderstorms. What is the probability that, in a
particular day of rainy season, it will rain and it will thunderstorm?

Solution:
Let,
R= it will rain on that particular day
T= it will thunderstorm on that particular day
Probability Laws: 42
Multiplication Law
Solution(contd.):
70 80
Here, given that, 𝑃 𝑅 = = 0.7 and 𝑃 𝑇|𝑅 = = 0.8
100 100

Therefore, the probability that, on that particular day of rainy

season, it will rain and it will thunderstorm is-

𝑃 𝑅 ∩ 𝑇 = 𝑃 𝑇|𝑅 𝑃 𝑅
T|R R
= 0.8 ∗ 0.7 = 0.56
Probability Laws: 43
Multiplication Law
Example 5:
Mr. Fahad and Mr. Khan has to tour abroad for their business frequently.
Mr. Fahad tours 65% of the times in a year at abroad and Mr. Khan tours
50% of the times in a year at abroad. What is the probability that, on
January 01, 2016, both Mr. Fahad and Mr. Khan will be at abroad?

Solution:
Let,
F= Mr. Fahad will be at abroad
K= Mr. Khan will be at abroad
Probability Laws: 44
Multiplication Law
Solution(contd.):
65 50
Here, given that, 𝑃 𝐹 = = 0.65 and 𝑃 𝐾 = = 0.5
100 100

Therefore, the probability that, on January 01, 2016, both Mr.

Fahad and Mr. Khan will be at abroad is-

K F
𝑃 𝐾∩𝐹 =𝑃 𝐾 𝑃 𝐹
= 0.5 ∗ 0.65 = 0.325 (K∩F)
Probability using contingency table 45

Example 7:
Below given a contingency table for Smoking status and Cancer status.
Smoking Status/ cancer Status Cancer Healthy Total
Smoker 7860 1530 9390
Non-smoker 5390 11580 16970
Total 13250 13110 26360
1. What is the probability that a randomly selected person is a smoker
2. What is the probability that a randomly selected person has cancer?
3. What is the probability that a randomly selected person is both
smoker and has cancer?
4. If a person is smoker, what is the probability that he also has cancer?
Probability using contingency table 46

Solution:
Let,
S= The person is smoker, N= The person is non-smoker
C= The person has cancer, H= The person is healthy

1. The probability that a randomly selected person is a smoker

9390
𝑃 𝑆 = = 0.356
26360
Probability using contingency table 47

Solution (contd.):
2. The probability that a randomly selected person has cancer
13250
𝑃 𝐶 = = 0.503
26360

3. The probability that a randomly selected person is both smoker and

has cancer
7860 7860 9390
𝑃 𝑆∩𝐶 = = 0.298 = 𝑃 𝐶 𝑆 𝑃 𝑆 = ×
26360 9390 26360
Probability using contingency table 48

Solution (contd.):
4. If a person is smoker, what is the probability that he also has cancer

7860
7860 𝑃(𝑆 ∩ 𝐶) 26360
𝑃 𝐶𝑆 = = 0.837 = =
9390 𝑃(𝑆) 9390
26360