BUSINESS STATISTICS:

Text and Problems

With Introduction to Business Analytics

Dr. N. D. VOHRA
Chapter 6

Theory of probability
 INTRODUCTION

• Uncertainty is an all-pervasive feature of management

decision-making.
• Examples:
❑ How the market would react to a new product to be
introduced by a company,
❑ whether more advertising will add to the product sales and
which mode of advertising will be the most desirable
❑ How the labor union will react to a new promotion policy to
be introduced by the management, etc.
 MEANING OF PROBABILITY

• Probability is a measure of the chance that an uncertain

event will occur.
• It takes on numerical values when considered in
relation to an experiment.
• An experiment in turn is a well-defined procedure or a
situation that results in one of the possible outcomes
 BASIC CONCEPTS

• An activity or measurement that

Experiment results in an outcome
• Example: Rolling of a Dice

• Possible outcome/s of experiment

Sample points • Example: Getting 2 on dice

• Set of all the possible outcomes of

Sample space an experiment
• Example: {1, 2 , 3, 4, 5, 6}
 Sample Space: Set of
all possible outcomes
of an experiment

Event: any set of possible outcomes in a

sample space
Elementary Compound
Events: contain a Events: contain
single sample more than one
point sample point.
 Classical Method

APPROACHES
TO
CALCULATION Relative Frequency
OF
PROBABILITY

Subjective
 CLASSICAL APPROACH

•
 RELATIVE FREQUENCY APPROACH

• Probability of an event can be obtained by

❑ Repeating the experiment over and over again,
❑ Under similar conditions,
❑ and taking the ratio of favourable outcomes to the total
number of trials.
• The probability assessment of event E is given by:
 PERSONALISTIC APPROACH

• Probability of an event is a measure of the degree of

belief that an investigator has in the happening of the
event.
• Example: On the question as to what are the chances
that a new product to be launched by a company will be
successful, different people may have different opinions.
Thus, whereas chances by one may be placed at 80
percent, another may put the chances to be 95 percent.
Accordingly, the two will assign probabilities 0.80 and
0.95 respectively to the event.
 MARGINAL, JOINT AND
CONDITIONAL PROBABILITY

• Marginal Probability refers to the probability of

occurrence of a simple event.
• Joint Probability is calculated when we are
interested in the occurrence of two or more events
simultaneously or in unison.
• Conditional Probability is the probability of an
event when some information about the events
involved is already given.
 EVENT RELATIONSHIPS

• Mutually Exclusive : if two or more events cannot occur

simultaneously.
❑ The events of an employee being late and absent on a given
day are mutually exclusive, since one cannot both be late and
absent simultaneously.
• Overlapping Events: if two or more events defined on a
sample space can occur severally and jointly, they are called
overlapping events.
❑ The events that a given TV set has a scratch and a defective
picture tube are not mutually exclusive and are overlapping
instead.
 MUTUALLY EXCLUSIVE AND
OVEERLAPPING EVENTS

Mutually Exclusive
Events
Overlapping Events
 EVENT RELATIONSHIPS

• Collectively Exhaustive Events: If the union (or

collection) of two or more events is equal to the
sample space on which they are defined, they are said
to be collectively exhaustive.
• Complementary Events: Two events, E and
(non-E) are said to be complementary when the
sample space is partitioned into two segments, one
that represents the occurrence of E and the other
that is not a part of E.
 COLLECTIVELY EXHAUSTIVE AND
COMPLEMENTARY EVENTS

Collectively Exhaustive
Events Complementary Events
 EVENT RELATIONSHIPS

• Independent and Non-independent Events:

Two or more events are said to be independent if the
occurrence or non- occurrence of one event does not
influence the occurrence of the other(s).
If it does, they are not independent.
 PROBABILITY THEOREMS

• These are applied when we deal with two or more

events.
• Theorems are:
❑ Theorem for Complementary Events
❑ Theorem of Addition
❑ Theorem of Multiplication
❑ Theorem of Total Probability
❑ Bayes’ Theorem
 PROBABILITY THEOREM FOR
COMPLEMENTARY EVENTS

•
 EXAMPLE

The probability of a piece of equipment to run

successfully during a month is 0.82. What is the
probability that it will fail during the month?

Solution:
The two events being complementary to each other,
P(equipment to fail) = 1 – P(equipment to function)
= 1 – 0.82 = 0.18
 THEOREM OF ADDITION

Used to calculate the probability of

occurrence of either of two or more events

When events are

When events are mutually
overlapping
exclusive

P(E1 ∪ E2) = P(E1) + P(E2) P(E1 ∪ E2) = P(E1) + P(E2) – P(E1∩ E2)
 EXAMPLE: THEOREM OF ADDITION

A box contains 20 discs numbered consecutively from 1 to 20. A

disc is selected at random. Find the probability that the number
on disc is divisible by 3 or 5.
Let E1 be the event that the number is divisible by 3, and E2 be the event
that it is divisible by 5.
E1 includes 3, 6, 9, 12, 15, and 18 E2 includes 5, 10, 15, and 20,
The number 15 is included in both the lists. The events are overlapping.
P(E1 ∪ E2) = P(E1) + P(E2) – P(E1 ∩ E2)
P(E1) = 6/20, P(E2) = 3/20, and P(E1 ∩ E2) = 1/20
P(E1 ∪ E2)= 6/20 + 3/20 – 1/20 = 8/20
 THEOREM OF MULTIPLICATION

Used to calculate the probability of joint occurrence of

both (all) of two (or more) events

When the events are When events are not

independent independent

P(E1 ∩ E2) = P(E1) × P(E2)

P(E1 ∩ E2) = P(E1) × P(E2/E1)
P(E1 ∩E2) = P(E2) × P(E1/E2)
 EXAMPLE:
THEOREM OF MULTIPLICATION

A mathematical problem is given to two students A and B. Their

chances of solving it correctly are known to be 0.4 and 0.9
respectively. Find the probability that both of them solve it.
The events E1 that A solves the problem, and E2 that B solves the
problem, are independent.
Hence,
P(E1 ∩ E2) = P(E1) × P(E2)
= 0.4 × 0.9 = 0.36
Thus, chances are 36 percent that both of them solve the problem.
 STATISTICAL INDEPENDENCE

• When two events E1 and E2 events are known to be

independent, then P(E1 ∩ E2) = P(E1) × P(E2).
• This probability relationship can also be used to test
independence of the events involved.
• If three probabilities P(E1 ∩ E2), P(E1) and P(E2) are given
then we can determine whether P(E1 ∩ E2) = P(E1) × P(E2).
• If the equation does hold true, the events E1 and E2 are termed
as statistically independent.
• If this relationship does not hold, the events said to be
statistically not independent.
 DIFFERENCE BETWEEN MUTUALLY
EXCLUSIVE AND INDEPENDENT EVENTS

• Mutually exclusive events cannot be independent.

• For two independent events with non-zero probabilities,
the joint probability is greater than zero but for mutually
exclusive the probability of joint occurrence is always
equal to zero.
• Independent events cannot be mutually exclusive.
• Mutually exclusive events are statistically dependent.
• However, statistically dependent events do not need to be
mutually exclusive.
 THEOREM OF TOTAL PROBABILITY

• If an event E is associated with other intermediate,

mutually exclusive events H1, H2, …., Hn, then the
total probability of its occurrence is given by
P(E) = P(H1 ∩ E) + P(H2 ∩ E) + …. +P(Hn ∩ E)
= P(H1) × P(E/ H1) + P(H2) × P(E/ H2) + …
+P(Hn) × P(E/ Hn)
 EXAMPLE:
THEOREM OF TOTAL PROBABILITY

Suppose the Personnel Manager of a company wants to introduce a

new policy of assessment of employees of the company for
promotion on their jobs which is subject to clearance from the
General Manager (GM).
The post of GM is vacant currently and likely to be filled soon by
appointment of one of the three deputy GMs, A, B or C. The
chances of A, B and C of being appointed as GM are 0.5, 0.3 and 0.2
respectively, while the likelihood of this policy implementation is 0.1,
0.6, and 0.9 respectively with them.
What is the probability that the policy will eventually be
implemented?
 EXAMPLE:
THEOREM OF TOTAL PROBABILITY

Let H1, H2 and H3 represent the respective events that A, B and C are
promoted, and E be the event that policy is implemented
P(H1) = 0.5, P(H2) = 0.3, P(H3) = 0.2, P(E/ H1) = 0.1, P(E/ H2) = 0.6 and
P(E/ H3) = 0.9.
P(E) = P(H1 ∩ E) + P(H2 ∩ E) + P(H3∩ E)
= P(H1)× P(E/H1) +P(H2) × P(E/H2) + P(H3) × P(E/H3)
Substituting the given values,
P(E) = 0.5 × 0.1 + 0.3 × 0.6 + 0.2 × 0.9 = 0.05 + 0.18 + 0.18 = 0.41
Thus, the given policy has an overall 41 percent chance of being introduced.
 PROBABILITY TREE

• Probability trees are a useful graphic procedure for

handling problems involving multiple levels of
conditional probabilities.
• A probability tree involves a series of branches
covering all possible outcomes of an experiment.
• The sum of all probabilities at the end of final
branches add up to 1.
 BAYES’ THEOREM

•
 BAYES’ THEOREM CALCULATIONS

• Inputs:
❑ Prior probabilities
❑ Conditional probabilities
• Steps:
❑ Calculate joint probabilities by multiplying prior and
conditional probabilities
❑ Add the joint probabilities to obtain total probability
❑ Take ratios of various joint probabilities and total
probability to get posterior probabilities
END OF CHAPTER 6