In general, a Venn diagram can be used to illustrate the sample space (s) and an event A.

The
number of outcomes in the sample space is denoted by n(s). The number of outcomes in
event A is denoted by n (A).

APPROACHES TO PROBABILITY

Three methods (or approaches) for determining the probability of an event can be identified
as:

(1) Classical (a priori)

(2) Empirical
(3) Subjective

The Classical Method—This theory is applicable when the range of possible uncertain
outcomes is known and equally likely, i.e., each simple event has the same probability of
occurring. Logic can thus be applied to determine the probabilities of each outcome.
n A
If the probability of an event A is denoted by P A , then P A = ,
n S
where n A = Number of equally likely outcomes associated with the event
n S = Total number of equally likely outcomes

The probability of not occurring is P not A 1 P A

Example 1
Find the probability that a head shows up when a coin is thrown.

Solution
Here, S = (heads, Tails) and A = (heads)
ns 2 and n A 1
n A 1
P A
ns 2

Example 2
In a class of 50 students, 20 are girls and the remaining 30 are boys. If a student is chosen at
random from this class, what is the probability that the student is (a) a boy (b) a girl?
Solution
Let the sample space be S, and
B = the event that a boy is chosen
G = the event that a girl is chosen

79
(a) ns 50 and n B 30
n A 30 3
PB
ns 50 5

(b) ns 50 and n G 20
nG 20 2
PG
ns 50 5

Example 3
If a die is thrown once, what is the probability that
(a) 2 or 3 shows up (b) an even number shows up (c) a prime number shows up?

Solution
Here S 1, 2, 3, 4, 5, 6 and n S = 6
(a) Let A be the event 2 or 3 shows up then n A = 2

n A 2 1
P A
ns 6 3
(b) Let B be the event an even number shows up. Then B 2, 4, 6 and n B = 3
nB 3 1
PB
ns 6 2

(c) Let C be the event a prime number shows up. Then C 2, 3, 5 and n C = 3
nC 3 1
PC
n A 6 2

Example 4
Three coins are thrown. What is the probability of;
(a)2 heads and 1 tail (b) all tails?

Solution
Here s HHH , HHT , HTH , THH , TTH , THT , HTT , TTT n s =8
(a) Let A be the event 2 heads and 1 tail
Then A HHT , HTH , THH and n A = 3
n A 3
P A
ns 8

(b) Let B be the event all tails

1
Then B TTT and n B = 1 PB
8

80
The Empirical Approach – This is applicable when we cannot rely on the exactness of a
process to determine the probabilities. In such situations, the probability of a given outcome
Z , P Z , is calculated as the ratio of the number of times that Z occurs divided by the

number of experiments, i.e. the number of times the experiment is conducted.

Mathematically,
No. of Z occurrences
P Z
No. of exp eriments

The Subjective Approach –Subjective probabilities are probabilities based upon an educated
guess. The approach simply defines probability as the strength of belief that an event will
occur.

Subjective Approach has many applications in business problems because the probabilities
cannot be derived from logic, nor are there sufficient empirical observations upon which to
base probability estimates e.g. forecasting of company profits by investment analysts, and
calculation of expected returns of financial investments.

PROBABILITY LIMITS

The probability of any event A occurring must be 0 and 1 inclusive. Symbolically, this can
be written as: 0 P( A) 1

(i) If P( A) 0 , then A is known as an impossible event. Examples include a negative

number showing up when a die is thrown, a student chosen at random from level 200
class being less than 10 years old.
(ii) If P( A) 1 , then A is known as a certain event. Examples include an event of a
woman giving birth to a male or female.
The sum of the probabilities of all possible outcomes of an experiment must total
exactly 1.
1 1
e.g. If a coin is tossed we must have that P H P T 1
2 2
If the complement of any event A is written as A and defined as “event A does not occur”,
then the probabilities of A or A must be 1. Thus P A PA 1 which implies that
PA 1 P A

81
Alternatively subtracting the probability of any event occurring from 1 will give the
probability of the event net occurring.

RULES OF ADDITION
Special Addition Rule
To apply the special addition rule, the events must be mutually exclusive.

Mutually Exclusive Events

Two or more events are called mutually exclusive events if the occurrence of any one of them
excludes the occurrence of the others. In other words, mutually exclusive events are such
that, if one occurs none of the other events can occur at the same time.
The special rule of addition states that the probability of one or the other events occurring
equals the sum of their probabilities.
Thus, if A and B are mutually exclusive events, then P A or B P A P B and
P A and B 0
E.g. If a die is thrown once, the event “an odd number shows up” and the event “an even
number shows up” are mutually exclusive events.
Thus given A = the event that an odd number shows up
B = the event that an even number shows up
3 3
Then P A or B P A P B 1
6 6
Example
A bag contained 8 red, 4 white and 6 blue balls, all of the same material and size. If a ball is
picked at random from the bag, what is the probability picking
(a) a red or a white
(b) a white or a blue ball?

Solution
Total number of balls 8 4 6 18
(a) Let be the event of picking red ball and be the event of picking a white ball
8 4
Then P R
18 9
4 2
and P W
18 9

82
 P R or W P R P W

4 2 6 2
& are mutually exclusive
9 9 9 3

(b) Let =the event of picking a White ball

and =the event of picking a Blue ball.
4 6
Then PW PB
18 18
4 6 10 5
P W or B = & are mutually exclusive
18 18 18 9
Note that, can be written as: P( A B) . A B is the union of events and , and
consists of those elements that are in either or or both.

Example
Of the 300 employees at TICO Ltd, 150 are classified junior workers; 110 are line workers
and the remaining 40 are senior workers. If an employee is selected at random, find the
probability that
(a) a junior worker or line worker is chosen
(b) a line worker or senior worker is chosen.

Solution
Let = event that a junior worker is chosen
= event that a line worker is chosen
= event that a senior worker is chosen

150 110
(a) P A and P B
300 300
Since A and B are mutually exclusive
P A or B P A P B

150 110 260 13

300 300 300 15

110 40
(b) PB and P C since B and C are mutually exclusive,
300 300
P B or C P B P C

110 40 150 1
300 300 300 2

83
THE GENERAL RULE OF ADDITION

Events may not be mutually exclusive e.g. customers at a bank may either have a Savings
Account, or Current Account or both. In such a situation, if we define A = event that a
customer has savings account and B = event that a customer has current account, then
A B .

That is, it is possible that a customer may have a savings account and a current account at the
same time. To find the joint probability of A or B we can apply the following formula:
P A or B P A P B P A and B

Example
Let P A 0.56 and P B 0.34 . If the probability that both occur is 0.20 what is the
probability of either A or B?

Solution
Given P A 0.56, P B 0.34 and P A and B 0.2 ,

Then P A or B P A P B P A B

0.56 0.34 0.2

0.70

RULES OF MULTIPLICATION
Special rule of multiplication: The special rule applies when the events are independent.

Independent Events
If the occurrence or non-occurrence of an event does not affect the probability of
occurrence of another event , we say that and are independent events. If and are
two independent events, then the probability that and occur i.e. P A and B can be

calculated as the product of and . Thus if and are any two independent events
of an experiment then P A and B P A P B orP A B P A P B .

Example
When a coin is tossed twice, what is the probability that the first shows up heads and the
second shows up heads?
Solution
Let = the event, 1st coin shows up heads

84
and = the event, 2nd coin shows up heads
and are independent events why?
1 1
Given P A and P B
2 2
1 1 1
P A and B P A P B
2 2 4
Example
If the probability of one customer making purchases on entering a shop is ½ and the
probability of another making purchases from the same shop is 2/3, what is the probability
that
(a) they both make purchases
(b) they both do not make purchases on entering the shop?

Solution
Let A =the event, the 1st customer makes purchases
and B=the event, the 2nd customer makes purchases
1 2
(a) The P A and P B
2 3
Since and are independent
P A and B P A P B

1 2 1
2 3 3
(b) 1st customer not making purchases is
1 1
PA 1 P A 1
2 2
2nd customer not making purchases is B
2 1
PB 1 PB 1
3 3

Hence P A and B P A P B

1 1 1
2 3 6

Example
The probability that in a week there is a power failure is 1/3 and that there will be a raw
materials shortage is 2/7. Find the probability of

85
(a) a power failure and a raw material shortage in a week
(b) no power failure and no raw materials shortage

Solution

Let the event a power failure

the event a raw material shortage.
1 2
(a) Then P( F ) and P( M )
3 7
P( F and M ) P( F ) P(M )
1 2 2
3 7 21

(b) Let F the event, no power failure

M the event, no raw material shortage
2 5
P( F ) and P( M )
3 7
Therefore,
P ( F and M ) P( F ) P ( M )
2 5 10
3 7 21

CONDITIONAL PROBABILITY
If and are two events, then the probability that occurs given that has occurred is
denoted by or and is called the conditional probability of given that
has occurred.

The conditional probability of given is defined as follows:

P A and B P A B
P B A or P B A
P A P A

Example
Given that, for the events and ,
n A 25, n B 15 and n A B 10

Calculate: (a) P A B (b) P B A

86
Solution
Show the data on a Venn diagram as below:

A(25) B(15)

15 10 5

Here n s 15 10 5 30
25 5
Therefore, P A
30 6
15 1
PB
30 2
10 1
and P A and B
30 3

P A and B 13 2
(a) P A B
P B 12 3

P A and B 13 6 2
(b) P B A
P A 56 15 5

Note that P A B PB A

Example
A group of twenty university students may be classified by sex and their major subject,
according to the table below:
Male Female
Commerce 8 2
Economics 4 6

If a student is chosen at random from the group, what is the probability that the student is a
female?
(a) given that the student takes Economics
(b) given that the student takes Commerce?

Solution

87
Let and be the events that a student takes Commerce and Economics respectively and M
and F the events that a student is a male and female respectively.
Then
(a) We seek P F E
10 6
Now P E ,P F E
20 20
P F E 6 20 6 3
P F E
P E 10 20 10 5

P F C 2 20 2 1
(b) P F C
P C 10 20 10 5

Example
A business employs 300 men and 200 women. Five percent (5%) of the men and 10% of the
women have been working there for more than 25 years. If an employee is selected by
chance, what is the probability that the employee is male, given that the length of
employment is more than 25 years?

Solution
Let =event an employee selected is a male
=event an employee selected is a female
=event an employee selected has worked there for over 25 years.

5 10
N B 300 200 35
100 100

Males who have worked for over 25 years 5 100 300 15

Females who have worked for over 25 years 10 100 200 20

No. of employees who have worked for 25 years 15 20 35

We seek P M B
35 15
But P B and P M B
500 500
P M B 15 500 15 3
P M B
P B 35 500 35 7

88
The Multiplication Rule

If the events are dependent, then, by definition, we must consider the first event in
determining the probability of the second. That is, the probability of event depends on the
condition that has already occurred. Applying the principle of conditional probability for
any two events and ;

P A B
P B A P A B P A P B A
P A

P A B
Also P A B P A B P B P AB
P B

Example
Two cards are drawn from a pack of 52 cards in succession (a) without replacement (b) with
replacement. Calculate the probability that they are both aces?
Solution
Let A be the event the 1st card drawn is an ace and B the event that the 2 nd card drawn is an
ace.
Then we seek P A B

(a) Since A and B are dependent,

P A B P A P B A
4 3 1
52 52 169

Example
A light bulb manufacturer produces bulbs in batches of 50 and it is known that 5 of each bath
will be defective. If 2 bulbs are selected
(a) with replacement
(b) without replacement
What is the probability that both are defective?

Solution
Let be the event the bulb selected is defective.
Then we seek P A1 A2
5 5
(a) P A1 and P A2
50 50

89
 P A1 A2 P A1 P A2
5 5 1
50 50 100

5 4
(b) P A1 and P A2 A1
50 49
P A1 A2 P A1 P A2 A1
5 4 2
or 0.00816
50 49 245

MULTIPLICATION THEOREM FOR 3 OR MORE EVENTS

If is any integer with n 3, and

A1 , A2 , A3 An are any events for which

P A1 A2 A3 An 1 0,
Then,
P A1 A2 A3 An P A1 P A2 A1 P A3 A1 A2 P A4 A1 A2 A3
P An A1 A2 An 1

Example
From a box containing 9 white balls and eleven (11) green balls, 5 balls are drawn at random,
one after the other without replacement. Find the probability that the balls in order drawn are
white, green, green, white, green.

Solution
Let Wi i 1, 2, 3, ,5 be the event that the ball drawn is white and G j j 1,2,3 5 be

the event that the ball drawn is green. We seek P W1 G2 G3 W4 G5 .

P W1 G2 G3 W4 G5 P W1 P G2 W1 P G3 W1 G2 P W4 W1 G2 G3
P G5 W1 G2 G3 W4

9 11 10 8 9
20 19 18 17 16
99
or 0.038
2584

90
THEOREM ON TOTAL PROBABILITIES

Definition:

The events A1 , A2 , A3 Am from a partition of the sample space (s) into m subsets if they are
mutually exclusive and is exhaustive, i.e. if

(i) P Ai Aj exclusiveness
m
(ii) S A exhaustiveness
i 1

Eg. S
A1
A4

Theorem:
Let be an event with respect to the sample space (s) and let A1 , A2 , A3 Am be partitions of

. If P Ai 0 where i 1, 2, 3, , m then
m
PB P Ai P B Ai
i 1

For example if , then

PB P Ai P B Ai P A1 P B A2 P A3 P B A3
m
m
Proof: S Ai B Bn i
i 1 i 1

A1nA2
m m
PB P BnAi P B Ai P Ai
i 1 i 1

Example
Three holiday caravans, X, Y, Z, are let to 2 adults and 4 children; 4 adults and 2 children and
3 adults and one child respectively. A special holiday price is awarded by first choosing a
caravan at random and then choosing an occupant at random. Find the probability that the

91
winner is
(a) a child (b) an adult.

Solution
Let , , and be the events caravans X, Y and Z are chosen respectively.
Then P X 1 3, P Y 1 3 andP Z 13
Also, let = the event a child is chosen
= the event an adult is chosen

(a) Then P C X 2 6, P C Y 2 6 and P C Z 14

PC P X PC X PY PC Y PZ PC Z
1 4 1 2 1 1
3 6 3 6 3 4
4 2 1 5
18 18 12 12

(b) P A X 2 6, P AY 4 6, P A Z 34
1 2 1 4 1 3
P A
3 6 3 6 3 4
2 4 3 7
18 18 12 12

Example

The management of a bank has decided to award any one employee from any of its two
branches A and B. Branch A has 10 employees, consisting of 6 males and 4 females. Branch
B also has 12 employees made up of 8 males and 4 females. A Branch is chosen at random
and then choosing an employee at random for the award. Find the probability that the award
winner is
(a) a male (b) a female.

Solution

Let A = the event branch A is chosen

B= the event branch B is chosen

P A 1 2 and P B 1 2 since both are equally likely to be chosen.

Also, let the event a male is chosen

the event a female is chosen

92
Then
(a) P M A 6 10 and P M B 8 12

P M P A P M A P B P M B
1 6 1 8
2 10 2 12
3 1 19
10 3 30
(b) P F A 4 10 and P F B 4 12

PF P A PF A PB PF B
1 4 1 4
2 10 2 12
1 1 11
5 6 30

BAYES’ THEOREM
Reverend Thomas Bayes, an 18th century English Presbyterian Minister developed a concept
useful in calculating curtain probabilities. The concept was developed when attempting to
develop a formula to arrive at the probability that God does exist based on evidence that was
available to him on earth. His work was later refined by Laplace who named it Bayes’
theorem. Bayes’ theorem is a formula which can be thought of as “reversing” condition
probability. That is, it finds a conditional probability given among other things, its
inverse .

Let B1 , B2 , B3 , , Bm be mutually exclusive events whose union is the sample space (s) of
an experiment.

Let A be an arbitrary event of such that P A 0 . Then for each i i 1, 2, 3, , m we

P Bi A P Bi P A Bi
have the probability of P Bi A m
P A
P Bi P A Bi
i 1

ILLUSTRATION OF BAYES THEOREM TREE DIAGRAM

Consider three possible outcomes, A, B and C with their respective conditional probabilities,
we can present them in a diagrammatic form as shown in Figure

93
DEDUCTION FROM THE TREE DIAGRAM

P(S) is equal to the sum of all the joint probabilities of S, i.e.

Similarly,

is equal to the sum of all the joint probabilities of , i.e.

The revised probabilities are given by , , ,

etc.

Example
The export department of a local manufacturing firm exports products to three different
markets, America, Asia and Europe. 40 percent of all manufactured goods were exported to

94
America, 30 percent to Europe and the rest, Asia. Thirty percent of products exported to
American and European markets were rejected and 1 percent in the Asian market.
i. What is the probability that a product selected at random from the export department
was rejected?
ii. If a product selected at random from the export department was rejected, what is the
probability that it was in the American market?

Example
Pozo Ltd uses two machines to manufacture its output. Machine A produces 60% of the total
output and machine B produces the remaining 40%. 3% of the units produced by A are
defective, while B has defect rate of 5%. A unit of output is selected at random and found to
be defective. What is the probability that is produced by?
(a) machine A? (b) machine B?

Solution
Let = the event a unit is produced by machine A
= the event a unit is produced by machine B
60 40
Then P A 0.6 and P B 0.4
100 100
Also, let be the event a unit selected is defective.
3 5
Then P D A 0.03 and P D B 0.05
100 100

(a) We seek P
AD .
P A D P A P D A
P A D
P D P A P D A P B P D B

0.6 0.03
0.6 0.03 0.4 0.05
0.018
0.47
0.018 0.02

(b) We seek P B D .

P B D P B P D B
P B D
P D P B P D B P A P D A

0.4 0.05
0.4 0.05 0.6 0.03

95
 0.02
0.53
0.02 0.018

Example
Three candidates are running for the office of president of a country. The probability that Mr.
A will be elected is 0.25, the probability that Mr. B will be elected is 0.55, and the probability
that Mr. C will elected is 0.2.

Should Mr. A be elected the probability that tuition fees will be introduced is 0.70; should
Mr. B or Mr. C be elected the corresponding probabilities for an introduction of tuition fees
are 0.10 and 0.40 respectively.

(a) What is the probability that tuition fees will be introduced after the election?
(b) If someone reads three months after the elections that tuition fees have been
introduced, what is the probability that Mr. B was elected?

Solution
Let , , and be the events that Mr. A, Mr. B and Mr. C are elected respectively.
Then, P A 0.25, P B 0.55 and P C 0.2

Also, let be the event that tuition fees will be introduced.

Then P X A 0.70, P X B 0.10 and P X C 0.40

(a) We seek P X .
Applying the total probability rule,
P X P A P X A PB PX B PC P X C
0.025 0.70 0.55 0.10 0.20 0.40
0.175 0.055 0.08
0.310

(b) We seek P B X
Applying the Bayes’ theorem
P B X P B P X B
P B X
P X P X

0.55 0.10
0.310

96
 0.177

REVIEW QUESTIONS

1. A bag contains 8 white balls, 5 red and 7 blue balls, all of the same material and size.
If a ball is picked at random, what is the probability of getting
(a) a white ball (b) a white or blue ball (c) a red or blue ball?

2. A coin is tossed thrice. What is the probability that

(a) 3 heads show up (b) 2 heads and 1 tail show up (c) at least one head
shows up?

3. Let A and B be two events defined on the same samples space (s) such that
1 1 1
P A ,P B and P AuB
3 3 2

4. The probability that a man will be alive in 25 years is 3/5 and the probability that his
wife will be alive in 25 years is 2/3. Find the probability that
(a) both will be alive (b) only the man will be alive (c) at least one
will be alive.

5. A firm has tendered for two contracts. It estimates that it has probability of 0.4 of
obtaining contact A and probability of 0.8 of loosing contract B. Find the probability
that the firm: (a) obtains both contracts (b) looses both contracts (c) obtains
exactly one contract.

6. The purchasing department of a company has analysed the number of orders placed
by each of the 5 departments in the company by type for this financial year as given in
the table below:
Order Type Sales Purchase Production Accounts Maintenance Total
Consumables 10 12 4 8 4 38
Equipment 1 3 9 1 1 15
Special 0 0 4 1 2 7
Total 11 15 17 10 7 60

97
 An error has been found in one of these orders. What is the probability that the
incorrect order:
(a) was equipment
(b) was consumables
(c) came from maintenance or production
(d) was an equipment order from production
(e) came from neither maintenance nor production?

7. A manufacturing company has plants in Accra and Tema. The Accra plant produces
40% of the total output, with a 10% defect rate. The Tema plant has a 20% defect
rate. If a single unit is found to be defective, what is the probability that it is produced
by
(a) Accra plant (b) Tema plant

8. There are three insurance companies A, Band C in a certain community. 20% of the
people in the community have an insurance policy with company A, 30% with
company B and 50% with company C. It is known from previous records that each
year 2%, 5% and 3% of those insured by companies A, B, and C respectively, submit
claims.
(a) What is the probability that a randomly selected person in the community
made a claim in a certain year?
(b) If a randomly selected person in the community has made a claim in a certain
year, what is the probability the person has a policy with
(i) insurance company A (ii) insurance company B?
PROBABILITY DISTRIBUTIONS

A probability distribution gives the probability of each possible value of the variable.
Consider this situation:

By mistake, 3 faulty fuses are put into a box containing two good fuses. The faulty and good
fuses become mixed up and are indistinguishable by sight. You take 2 fuses from the box.
What is the probability that you take?
(a) no faulty fuses
(b) none faulty fuse
(c) two faulty fuses

It is possible to show the outcomes and probabilities on a tree diagram:

98
 Probability Outcome
F P(F, F) = 2 x 2 = 0.3 2 faulty fuses
2/ 5 4

3/ F P(F, G) = 3 x 2 = 0.3 1 faulty fuse

2/ G 5 4
3/ F
2/ G P(G, F) = 2 x 2 = 0.3 1 faulty fuse
5 4
1/
G
P(G,G) = 2 x 1 = 0.1 0 faulty fuses
5 4

(a) P (no faulty fuses) = 0.1

(b) P (one faulty fuse) 0.3 0.3 0.6
(c) P (two faulty fuses) = 0.3

The variable being considered here is “the no of faulty fuses” and it can be denoted by x.
The values X can take are 0, 1 or 2.

The probability that there are no faulty fuses, i.e. the probability that the variable X takes the
value 0, can be written P X 0 , so P X 0 0.1

Similarly P X 1 0.6 and P X 2 0.3

When defining variables, the variable is usually denoted by a capital letter (X,Y,R etc) and
a particular value that the variable takes by a small letter (x, y, r, etc) so that P(X = x) means
“the probability that the variable X takes the value x”.

The probability distribution for X can be summarized in a table

X 0 1 2
P (X = x) 0.1 0.6 0.3

If the sum of the probabilities is 1, the variable is said to be random. In this example,
P X 0 P X 1 PX 2 0.1 0.6 0.3 1
So X is a discrete random variable.

Theorem

99
For a discrete random variable, the sum of the probabilities is 1. The fn that is responsible for
allocating probabilities P X x is known as the probability density function of X,
sometimes abbreviated to the p.d.f of x. The probability density fn can either list the
probabilities individually or summarise them in a formula.

Example
The discrete random variable W has probability distribution as shown.

W -3 -2 -1 0 1
P (W = w) 0.1 0.25 0.3 0.15 d

Find
(a) the value of d, (b) P 3 W 0 (c) P W 1 (d) P 1 W 1
(e) the mode

Solution
(a) Since PW w 1
allw

0.1 0.25 0.3 0.15 d 1

0.8 d 1 d 0.2
(b) P 3 W 0 PW 3 PW 3 PW 2 PW 1
0.1 0.25 0.3 0.65
(c) PW 1 PW 0 PW 1
0.5 0.2 0.35

(d) P 1 W 1 PW 0 0.15

(e) The value of w with the highest probability is –1, so the mode = 1.

Exercise
1. The discrete random variable x has the given probability distribution
X 1 2 3 4 5
P (X = x) 0.2 0.25 0.4 0 0.05

(a) Find the value of a

(b) Find (i) P 1 X 3 , (ii) P X 2 (iii) P 2 X 5

100
THE BINOMIAL PROBABILITY DISTRIBUTION

The binomial probability distribution is one of the most widely used discrete probability
distributions. It is applied to find the probability that an outcome will occur x times in n trials
of an experiment.

For example, suppose that sales records of a supermarket indicate that 40% of all customers
who enter the shop make a purchase. What is the probability that 2 of the next 3 customers
will make a purchase?

In order to find this probability, we first note that the experiment of observing three
customers making a purchase decision has several distinguishing characteristics.

1. The experiment consists of a sequence of three identical trials, where each trial
consists of a customer making a purchase decision.
2. Two outcomes are possible on each trial: the customer makes a purchase (which is
called success) or the customer does not make a purchase (which is called a failure).
3. Since 40% of all customers make a purchase, it is reasonable to assume that
is 0.4 and is constant for all customers and that is 0.6 which
is also constant for all customers.

Denoting success by A and failure by B, the possible outcomes of the experiment will be as
follows:

It can be seen that the sample space of the experiment consists of eight outcomes. Here, for
example, the outcome AAA indicates all 3 customers making purchases. On the other hand,
the outcome AAB indicates the first customer making a purchase, the second customer
making a purchase and the third not making a purchase.

Two out of three customers make a purchase if one of the outcomes, occurs.
Since the trials (purchase decisions) are independent, we can multiply the probabilities
associated with the different trial outcomes (each of which is or ) to find the probability
of a sequence of outcomes.

It follows that the probability that two of the next three customers make a purchase is
2 2 2 2
04 0. 6 0.4 0.6 0. 4 0. 6 3 0.4 0.6 0.288

101
Note that each of the outcomes consists of two successes and one failure and the probability
2 1
of a each of those outcomes is 0.04 0. 6 p 2 q1 . Where P is the probability of success and
q is the probability of failure. Here p is raised to a power that equals the number of successes
(2) in the three trials, and q is raised to a power that equals the number of failures in the three
trials.

In general, each of the outcomes describing the occurrence of successes in n trials

represents a different arrangement of successes in n trials. However, each of these
outcomes consists of successes and n - failures. Therefore, the probability of each
outcome is p x q n x . Hence, the probability that of the next trials are successes is

n n!
Cx p x qn x
where n
Cx
x !(n x)! .
This is the binomial model.

Conditions for a Binomial Model

For a situation to be described using a binomial model, the following four conditions must be
satisfied:
1. There are n identical trials (i.e fixed no. of trials)
2. The outcome of each trial is deemed either a success or a failure (i.e. each trial has
only two outcomes)
3. The probability p of a successful outcome is constant for each trial
4. The trials are independent.

The discrete random variable, , is the number of successful outcomes in n trials. If the above
conditions are satisfied, is said to follow a binomial distribution. This is written
X ~ B n, p or X ~ Bin n, p ~ Bin (n, p).

NB: The number of trials and the probability of success , are both needed to describe
the distribution completely. They are known as the parameters of the binomial distribution.
Writing as q , q 1 p

If X ~ B n, p , the probability of obtaining successes in trials is where

P x r nCr p r q n r for r 0, 1, 2, 3, ,n

Example 1

102
In a particular population, 10% of the people are above 190cm. If three people are selected at
random from the population, what is the probability that
(a) exactly 2 of them are above 190cm?
(b) all three are above 190cm?
(c) none is above 190cm?

Solution
When 3 people are selected, n 3, p 0.1 and q 1 0.1 0.9
Let = number of successful outcomes in 3 trials
So X ~ B 3,0.1
3
(a) p x 2 C2 p2q1 3 0.1 0.9 0.072
3 3
(b) px 3 C3 p3q0 p3 0.3 0.001
3 3
(c) px 0 C0 p 0 q 3 p3 0.3 0.729

Example 2
In the mass production of belts it is found that 5% are defective. Belts are selected at random
and put into packets of 10. A packet is selected at random. Find the probability that is
contains
(a) 3 defective belts
(b) less than 3 defective belts

Solution
Here, and
So X ~ B 10,0.5
10
(a) p x 3 C3 p 3q 7
3 7
120 0.05 0.95
120(0.000125)(0.698337296)
0.0105
(b) p ( x < 3) = p (x = 0) + p(x = 1) + p( x = 2)
10
= C 0 p 0 q10 +10 C1 p1q 9 + 10 C 2 p 2 q 8
= (0.05) 0 (0.95)10 + 10 (0.05)1 (0.95)9 + 45 (0.05) 2 (0.95)8
= 0.5987 + 0.3151 + 0.0746
= 0.9884

103
Example 3
At a bank, 60% of customers use ATM card. Find the probabilities that in a randomly
selected sample of ten customers,
(a) exactly two use ATM card.
(b) more than seven use ATM card
(c) none use ATM card

Solution 3
Let be the number of customers in a sample of 10 who use ATM card.
Then n 10, p 0.6 and q 1 0.6 0.4

X ~ B 10, 0.6

10 2 8
(a) P X 2 C2 p 2 q 8 45 0.6 0.4

(b ) p ( 7) p ( x 8) p( x 9) p ( x 10)
10 8 2 10 9 1 10
C8 p q C9 p q C10 p10 q 0
45(0.6)8 (0.4) 2 10(0.6) 9 (0.4) (0.6)10 (0.4) 0
0.17(2 s. f .)
10 0 10
(c) Px 0 = C 0 p 0 q10 0.6 0.4
= 0.00001

Example
The random variable is distributed
Find, correct to three decimal places;
(a) Px 3 (b) P1 x 4 (c) Px 1

Solution
P 0.2, q 1 0.2 0.8, n 7
4 3
(a) Px 3 = 7 C3 q 4q3 0.8 0.2
= 0.115 3d.p
(b) P1 x 4 =px 2 px 3 px 4
7 5 2 7 4 3 7 3 4
= c2 q p c3 q p c4 q p
5 2 4 3 3 4
21 0.8 0.2 35 0.8 0. 2 35 0.8 0. 2
0.419 3d . p

(c) Px 1 Px 2 Px 3 Px 7

104
 Rather than calculate all these terms, it is much quicker to find
P x 1 1 Px 1
1 Px 0 PX 1

1 q7 7
c1 q 6 p
6
1 0.8 7 7 0.8 0.2
0.423

Exercises
1. Five independent trials of an experiment are carried out. The probability of a
successful outcome is p and the probability of failure is q 1 – p.
Write out the probability distribution of x, where x is the number of successful
outcomes in five trials.

2. In an examination 60% of the candidates pass. Use the binomial distribution to

calculate the probability that a random sample of 10 candidates should
(a) two failures
(b) at most two failures

3. In a particular population, 10% of the people have blood type A. If four people what
is the probability that
(a) exactly two of them have blood type A?
(b) less than three people have blood type A
(c) at least one has blood type A

4. Based on experience, 50% of components produced by a plant are defective. Find the
probability that out of six components selected at random from the plant,
(a) exactly three are defective
(b) at least one is defective
(c) less than two are defective.

5. 30% of pupils in a school travel to school by bus. From a sample of a10 pupils
chosen at random, find the probability that
(a) only three travel to school by bus
(b) less than half travel by bus.

105
THE POISSON PROBABILITY DISTRIBUTION
The Poisson probability distribution, named after the French mathematician Simeon D.
Poisson, is another important probability distribution of a discrete random variable that has a
large number of applications.

THE POISSON DISTRIBUTION

The Poisson distribution is useful in describing a random variable that counts the number of
occurrences in particular interval of time: the number of equipment failures per week, the
number of traffic accidents per month, etc.

The Poisson distribution is a good approximation for the binomial when the number of trials,
n, is large n 20 and the probability of success (the probability of an occurrence in this

context) is small P 0.05

Poisson distribution:
u re u
P x r for x 0, 1, 2, 3,
r!
Where u is the mean of the distribution (and also happens to be the variance) and e is the base
of the natural logarithms e 2.71828

Poisson Assumptions
1. The probability that an event will occur in a short interval of time (or space) is
proportional to the size of the interval.
2. In a very small interval, the probability that two events will occur is close to zero.
3. The probability that any number of events will occur in a given interval is
independent of where the interval begins.
4. The probability of any number of events occurring over a given interval is
independent of the number of events that occurred prior to the interval.

Conditions to Apply Poisson Probability Distribution

The following three conditions must be satisfied to apply the Poisson probability distribution.
(a) is a discrete random variable
(b) the occurrences are random
(c) the occurrences are independent

106
The following are examples of discrete random variables for which occurrences are random
and independent and to which the Poisson probability distribution can be applied:

1. The number of customers arriving at a business premises during a one-hour interval

2. The number of defective components in the next 50 components manufactured by a
machine.
3. The number of cars sold by a car dealer in a given week or month.
4. The number of accidents that occur at a company during a one-month period
5. The number of customers making withdrawals from a bank during a one-hour
interval.

Example
It has been observed that for the past 100 hours customers entered our business. What is the
probability that at
a. exactly five customers will enter our business during the next hour
b. less than customers.

Solution
1000
This follows a Poisson distribution with u 10
10
(a) We seek P x 5
ure u 85 e 10
Px 5
r! 5!
0.0916

(b) We seek P x 3
Px 3 Px 0 Px 1 Px 2
0 8 1 8 2 8
8 e 8 e 8 e
0! 1! 2!
0.0003 0.0027 0.0107
0.0137

Example 2
The automatic teller machine (ATM) installed outside Ghana Commercial Bank Ltd, UCC
branch is used on average by 6 customers per hour. The bank closed this ATM for one hour
for repairs. What is the probability that during that hour
(a) eight customers came to use this ATM
(b) not more than three customers came to use this ATM.

107
Solution
Let u be the mean number of customers who use this ATM per hour. Then u =6
Let be the number of customers who came to use this ATM during the one-hour period
when it was closed for repairs. Then
(a) We seek P x 8
Substituting these values, we get
ure u 68 e 6
Px 8
r! 8!
0.1033

(b) Px 3 Px 0 Px 1 Px 2 Px 3

60 e 6 61 e 6 62 e 6 63 e 6
0! 1! 2! 3!
0.0025 0.0149 0.0446 0.892
0.1512

It must be noted that a Poisson interval can be adjusted provided the mean is adjusted
accordingly.

Example 3
It is observed that for the past 50 weeks, 200 customers of a bank defaulted in the repayment
of their instalment loan. What is the probability that
(a) for the coming week exactly 8 customers will default in repayment
(b) for the next two weeks fewer than 3 will default in repayment?

Solution
200
Let the mean number of defaulters per week be u, then u 4
50
Let be the number of customers that will default in repayment their instalment.
Then (a) We seek Px 8

ure u
Substituting, P x 8
r!
4 e4
8

8!
0.0298

108
 (b) Here, the interval has changed to two weeks. Thus the mean must be adjusted
to 2 4 8
Then we seek Px 3
Px 3 Px 0 Px 1 Px 2

80 e 8 81 e 8 82 e 8
0! 1! 2!
0.003 0.0027 0.0107
0.0137

Poisson Approximation to the Binomial

For a given binomial situation, if n is large n 30 and P is small (less than 0.1) the Poisson
distribution can be used as an approximation to the binomial distribution for calculating
probabilities.

When using the Poisson as an approximation to a binomial random variable, we set .

Example
The mortality rate of a certain disease is 5 deaths per 1000 cases. In a group of 400 cases of
the disease, what is the probability of (a) exactly 4 deaths (b) 3 or more deaths.

Solution
5 5
Here P ,n 400 & u np 400 2
1000 1000
Let the mean number of deaths be
5
U np 400 2
1000
Let be the number of deaths in the group.
We seek (a) P x 4 (b) P x 3

24 e 2
(a) Substituting, P x 4
4!
0.0902

109
(b) Px 3 1 Px 0 Px 1 Px 2 Px 3

20 e 2 21 e 2 22 e 2
1
0! 1! 2!

1 0.353 0.2707 0.2707

1 0.6767
0.3233

REVIEW QUESTIONS
1. A telephone switchboard handles on the average 4 calls per minutes. If the calls
follow a Poisson distribution, what is the probability of
(a) exactly 8 calls per minute?
(b) More than 10 calls per minute?

2. Items produced from a plant are known to be 2% defective. If the items are boxed
into lots of 100, what is the find a that a single box has
(a) exactly 4 defective
(b) 2 or more defectives?

3. The average number of new strikes in a company in Ghana has been observed to be 1
per week for the period 1992 – 2003. What is the probability that during any one
week
(a) there will be no new strike
(b) there will be 3 new strikes or more?

4. A manufacturing process used to make parts for another plant suffers a defect rate of
3%. The parts are transferred in lots of 200.
(a) Find the probability that a lot transferred contains exactly 5 defective parts.
(b) If more than 3 defective parts are detected the lot is rejected. Find the
probability that a lot transferred is rejected.

5. On average 3 new accounts are opened per day at the SSB Bank Ltd, CapeCoast
branch. Find the probability that on a given day the number of new accounts opened
at this bank will be

110
 (a) exactly 6
(b) at most 3

6. A sales person sells an average of 2 cars per day. Find the probability that
(a) on a particular day exactly 5 cars are sold by the sales person
(b) more than 2 cars are sold in 2 days.

111
THE NORMAL DISTRIBUTION

The normal distribution is one of the many probability distributions that a continuous random
variable can possess.

It is used to reflect the distribution of variables such as heights, weights, distances, and other
measurements that are infinitely divisible. Such continuous variables are generally the results
of measurement.

The Normal Probability Distributions

The normal probability distribution and its accompanying normal curve have the following
characteristics:

1. The normal curve is bell-shaped and has a single peak at the centre of the distribution.
2. The normal probability distribution is symmetrical about its mean. In other words,
one half of the curve is a mirror-image of the other.
3. The normal curve is asymptotic, meaning the two tails of the curve extend indefinitely
4. The total area under the curve is 1.0 or 100%.

The mean and the standard deviation or are the parameters of the normal distribution.
Given the values of these two parameters, we can find the area under a normal curve for any
interval.
It must be noted that each different set of values of and or gives a different normal
distribution. The value of u determines the centre of a normal distribution on the horizontal
axis and the value of or gives the spread of the normal distribution.

The Standard Normal Distribution

A number of normal distributions can occur in practice making it physically impossible to

provide a table of probabilities (such as for the binomial and Poisson) for each combination
of and . Fortunately, one member of the family of normal distribution can be used for all
problems where normal distribution can be used for applicable. This is the standard normal
distribution. The standard normal distribution is a special case of the normal distribution with
and .

Any normal distribution can be converted into the “standard normal distribution” by
subtracting the mean from each observation and dividing by the standard deviation. (Figure
1.1 displays the standard normal distribution curve)

112
Figure 1: Standard Normal Distribution Curve

The random variable that possesses the standard normal distribution is denoted by . In other
words, the units for the standard normal distribution curve are denoted by and variably
called, -value, -score, a -statistic, standard normal deviate or just normal deviate.
x u
The conversion formula or is given by Z
where:
is the value of any particular observation or measurement
is the mean of the distribution
is the standard deviation of the distribution

The -valueor -score is defined as “the number of standard deviations an observation lies
from the mean”.

Calculating Probabilities with the Score

Standardizing a normal distribution allows us to more easily determine the probability that a
certain event will occur. After determining the Z value using the above conversion formula
we can find the area or the probability under any normal curve by making reference to
appendix…

Example

113
Suppose we computed to be 1.98. What is the area under the normal curve between the
mean and X?

Solution
Using the table, go down the column of headed by to 1.9. Then move horizontally to the
right and read the probability under the column headed 0.08. It is 0.4761. This means that
47.61% of the area under the curve is between the mean and the value 1.98 standard
deviations above the mean. This is the probability that an observation is between 0 and 1.98
standard deviations of the mean.

Example 2
What is the area under the curve between the mean and for the following Z values?
(a) 1.11 (b) 1.65 (c) 1.96 (d) 2.58 (e) 2.84 (f) 3.56

Solution
Z –Value Area under Curve
(a) 1.11 0.3665
(b) 1.65 0.4505
(c) 1.96 0.4750
(d) 2.58 0.4951
(e) 2.84 0.4977
(f) 3.56 0.4998

Example
If a normal distribution of a continuous variable has a mean of 21.5 and standard deviation of
3.10, find the probability that a number selected at random (a) is greater than 25
(b) is less than 15 (c) is 22 and 30

Solution
Here u 21.5 and 3.10
Let be a random variable
Then we seek (a) P x 25 (b) P x 15 and (c) P 22 x 30

(a) The Z associated with 25 is

x u 25 21.5
Z 1.13
3.10

114
 P x 25 P z 1.13
0.5000 0.3708
0.1292

15 21.5
(b) Z 2. 1
3.10
P x 15 P z 2.1

P z 2.1 because the curve is symmetrical

0.5 0.4821
0.0179
22 21.5
(c) Z score for 22 is Z 0.16
3.10
30 21.5
Z score for 30 is Z 2.74
3.10
P 22 x 30 P 0.16 z 2.74

P z 2.74 P z 0.16
0.4969 0.0636
0.4333

Example
The weekly incomes of a large group of junior staff are normally distributed with a mean of
GH¢400,000 and standard deviation of GH¢40,000.
If a junior staff from this group is chosen at random, what is the probability that the weekly
income is
(a) greater than GH¢500,000
(b) less than GH¢300,000
(c) between GH¢250,000 and GH¢450,000?

Solution
= GH¢400,000 and GH¢40,000
Let = weekly income of a junior staff.
Then is normally distributed with
Then we seek (a) P x 500,000 (b) P x 300,000

P 250,000 x 450,000

115
 500,000 400,000
Z 2.5
40,000
P x 500,000 Pz 2.5
0.5000 0.4938 0.0062

300,000 400,000
(a) Z 2. 5
40,000
P x 300,000 Pz 2.5

P z 2.5
0.5000 0.4938 0.0062
250,000 400,000
(b) Z value corresponding to 250,000 is Z 3.75
40,000
450,000 400,000
Z value corresponding to 450,000 is Z 1.25
40,000
P 250,000 x 450,000 P 3.75 z 1.25

P z 3.75 P z 1.25
0.4999 0.3944
0.8943

The Normal Approximation to the Binomial

The binomial formula, which gives the probability of successes in trials is

n
P x z pr q n r

The use of the binomial formula becomes laborious and time consuming when n is large. For
example, if a problem involves taking a sample of 50, it becomes very difficult generating a
binomial distribution for that9 large number. A more efficient approach is to apply the
normal approximation to the binomial.

The approximation obtained by using the normal distribution is very close to the exact
probability when is large and is very close to .

Usually, the normal distribution is used as an approximation to the binomial distribution

when and are both greater than 5, that is, when and

116
Using the normal distribution as an approximation to the binomial involves the following
three steps:

Step 1: Compute u and for the binomial distribution, where u np and p npq
Step 2: Convert the discrete random variable to a continuous random variable. This is
done by making the correction for continuity.

Continuity Correction Factor

The value 0.5 is added or subtracted, depending on the problem, to/from a selected value
when a discrete probability distribution is being approximated by a continuous probability
distribution.

How to Apply the Correction Factor

Four situations may arise where there may be the need to add or subtract the correction
factor. The situations are

1. For the probability at least occur, use the area above x 0.5
2. For the probability that more than occur, use the area above x 0.5
3. For the probability that or fewer occur, use the area below x 0.5
4. For the probability that fewer than occur, use the area below x 0.6

Example
A coin is tossed 100 times. Use the normal curve approximation to find the probability of
obtaining.
(a) exactly 50 heads
(b) 60 or more heads

Solution
Let = no. of heads obtained
This is a binomial distribution with parameters
and
Since is large and p 0.5 , we use the normal approximation to the binomial distribution

with parameters U np 100 1 50

2

npq 100 0.5 0.5 5

117
(a) We seek P x 50
Applying the correction factor for continuity,
P x 50 P 49.5 x 50.5
49.5 50
For x 49.5; Z 0. 1
5
50.5 50
For x 50.5; Z 0.1
5
P 49.5 50.5 P 0.1 0.1
2P Z 1
2 0.0398 0.0796

(b) We seek P x 60 P x 59.5

59.5 50
For x 59.5; Z 1.90
5
P x 50.5 P z 1.90
0.500 0.4713
0.0287

Example
It has been observed that 50% of the customers of a bank use ATM card. If a random sample
of 30 customers is taken, what is the probability that
(a) 19 or more of them use ATM card
(b) Fewer than 18 use ATM card

Solution
Let be the number of customers that use ATM card.
This is a binomial distribution with parameters and
Since is large and , the normal approximation to the binomial can be used with
parameters: U np 30 0.5 15

npq 30 0.5 0.5 2.74

(a) We seek P x 19 P x 18.5

18.5 15
For x 18.5, Z 1.28
2.74
P x 18.5 P z 1.28

118
 0.500 0.3997 0.1003

(b) We seek P x 18 P x 17.5

17.5 15
For x 17.5, Z 0.91
2.74
P x 17.5 P z 0.91
0.500 0.3186 0.8186

REVIEW QUESTIONS

1. For a binomial probability distribution, n 150 and p 0.5 . Let be the number of
successes in 150 trials.
(a) Find the mean and standard deviation of the binomial distribution
(b) Find P x 80 using the normal approximation
(c) Find P 70 x 80

2. Find the following binomial probabilities using the normal approximation

(a) n 120 p 0.55 P x 70
(b) n 200 p 0.60 P 110 x 135
(c) n 150 p 0.40 P x 80
(d) n 150 p 0.40 Px 50

3. According to a survey 60% of cars imported into the country are at least 8 years in
2003. Assume that this result holds true for the cars imported currently into the
country. Find the probability that in a random sample of 200 cars at Tema harbour.
(a) exactly 110 are at least 8 years
(b) 130 or more are at least 8 years
(c) 115 to 135 are at least 8 years

MATHEMATICAL EXPECTATION

In probability theory, the Expected Value of a random variable is intuitively the long-run
average value of repetitions of the experiment it represents. For example, the expected value
of a die roll is 3.5 because, roughly speaking, the average of an extremely large number of
dice rolls is practically always nearly equal to 3.5. Less roughly, the law of large
numbers guarantees that the arithmetic mean of the values almost surely converges to the

119
expected value as the number of repetitions goes to infinity. The expected value is also
known as the Expectation, Mathematical Expectation, EV, Mean, or First Moment.

If is the probability that a person will receive a sum of money, , the mathematical
expectation, or simply the expectation, is defined as .

Example 1
If the probability that a man wins a GH¢5,000,000 prize is 1/5, his expectation is given as:
1 5,000,000 GH ¢ 1,000,000
5
Example 2
What is the expectation of a person who is to receive GH¢16,000,000 if he obtains 3 heads in
a single toss of 3 coins?

Solution
In a single trial, if H stands for head and T for tail, the following possibilities exist:
HHH , HHT , HTH , THH , HTT , THT , TTH , TTT
P HHH 18 P

The expectation is 1 8 16,000,000 GH ¢ 2,000,000

The concept of expectation can easily be extended.
If denotes a discrete random variable which can assume the values X 1 , X 2 , X 3 , , X K with

respective probabilities P1 , P2 , P3 , , PK where P1 P2 P3 PK 1, the expectation of

denoted by is defined as E X P1 X 1 P2 X 2 P3 X 3 PK X K PX

Example 3
A game is played with a fair die. The stake for each turn is GH¢4000. The prize is the
number of spots on the die multiplied by GH¢1000. Find the expectation of gain.

120
Solution
The probability of winning a prize of GH¢1000 is 1/6, as it is of winning GH¢2000,
GH¢3000, GH¢4000, GH¢5000 and GH¢6000. The stake money can be regarded as a
deduction of GH¢4000 from each prize.
1 1 1 1
E gain 1000 4000 2000 4000 3000 4000 4000 4000
6 6 6 6
1 1
5000 4000 6000 4000
6 6
1
3000 2000 1000 0 1000 2000
6
GH ¢ 500
Hence, in the long run, a player will lose GH¢500 per throw.

Example
A man is to enter into a business venture which can make a profit of GH¢6,000,000 with
probability 3/5 and take a loss of GH¢2,000,000 with probability 2/5. What is his
expectation?

Solution
3 2
Expectation 6,000,000 2,000,000
5 5
3,600,000 800,000

GH ¢ 2,800,000

Example 5
A contractor has to choose between two jobs. The first job promises a profit of
GH¢24,000,000 with a probability of 0.75 or a loss of GH¢6,000,000 (due to strikes and the
other delays) with a probability of 0.25.
The second job promises a profit of GH¢36,000,000 with probability of 0.50 or a loss of
GH¢9,000,000 with a probability of 0.50.
Which job should the contractor choose if he wants to maximise his expected profits and
why?

Solution
First Job:
E (Profit) 0.75 24,000,000 0.25 6,000,000

121
 18,000,000 1,500,000

GH ¢ 16,500,000

Expected profit from first job is GH¢16,500,000

Second Job
E (Profit) 0.50 36,000,000 0.50 9,000,000
18,000,000 4,500,000

GH ¢ 13,500,000

Expected profit from 2nd job is GH¢13,500,000. The first job should be chosen since
it has a higher expected profit.

EXPECTATION (MEAN)
More practically, the expected value of a discrete random variable is the probability-weighted
average of all possible values. In other words, each possible value the random variable can
assume is multiplied by its probability of occurring, and the resulting products are summed to
produce the expected value. The same works for continuous random variables, except the
sum is replaced by an integral and the probabilities by probability densities. The formal
definition subsumes both of these and also works for distributions which are neither discrete
nor continuous: the expected value of a random variable is the integral of the random variable
with respect to its probability measure.
n
E ( x) X i Pi
i 1

The expected value is a key aspect of how one characterizes a probability distribution; it is
one type of location parameter. By contrast, the variance is a measure of dispersion of the
possible values of the random variable around the expected value. The variance itself is
defined in terms of two expectations: it is the expected value of the squared deviation of the
variable's value from the variable's expected value. The variance and standard deviation could
therefore be computed using the formulae below:
n
2 2 2
Xi P Xi where variance
i 1

Standard deviation = var iance

n
2
A xi P Xi
i 1

122
EXPECTED VALUE MODELS— EMV & EOL
Once a probability distribution has been assessed for each set of uncertain states of nature—
and this can always be done, subjectively — it is straightforward to apply the next
step: compute the expected value for each action alternative. Since there are two ways to look
at the same problem (actual monetary values and opportunity losses), we can compute the
expected values on either one of the payoff tables.

EXPECTED MONETARY VALUE

In statistics, Expected Monetary Value (EMV) is the weighted average of probable outcomes.
It represents the expected average payoff if you made that decision using the same payoffs
and probabilities an infinite number of times.

That is, the total of the weighted outcomes (payoffs) associated with a decision,
the weights reflecting the probabilities of the alternative events that produce the
possible payoff. It is expressed mathematically as the product of an event's probability of
occurrence and the gain or loss that will result.

Referring to the original payoff matrix, the formula for expected monetary value (EMV) is:
n
EMV j Ei X ij Pi
i 1

where, EMV j = expected monetary value of action

X ij = payoff that occurs when the course of action is selected and event
occurs
Pi = probability of occurrence of event
= number of events
Criterion: Select the course of action with the largest EMV

EXPECTED OPPORTUNITY LOSS

An opportunity loss is the payoff difference between the best possible outcome under Sj and
the actual outcome resulting from choosing Ai given that Sj occurs.
In statistics, the opposite of an Expected Monetary Value (EMV) is the Expected Opportunity
Loss (EOL). An opportunity loss is the difference between the payoff for the decision you
made and the payoff you would have received if you had made the best decision. The EOL is
the average opportunity loss if you made an infinite number of decisions using the same
payoffs and probabilities.

123
Referring now to the opportunity loss matrix, the formula for expected opportunity loss
(EOL) is:
n
EOL( j ) Ei Lij Pi
i 1

Where,
Lij = is the difference between the value of the optimum action and the actual

values of the event

opportunity loss that
occurs when the course of action is selected and event occurs
Pi = probability of occurrence of event

EOL j = Expected opportunity loss

Criterion: Select the course of action with the smallest EOL.

Obviously, the same probability distribution applies (since the states of nature are the same).

Note that for a given probability distribution, the expected payoffs (EMV and EOL) for every
action alternative Ai always add up to a constant.

RETURN-TO-RISK RATIO (RTRR)

A ratio used by many investors to compare the expected returns of an investment to the
amount of risk undertaken to capture these returns. This ratio is calculated mathematically by
dividing the amount of profit the trader expects make when the position is closed (i.e. the
reward) by the amount he or she stands to lose if the price moves in the unexpected direction
(i.e. the risk). In other words, this ration expresses the expected monetary value as a ratio of
the standard deviation of an investment;
EMV j
RTRR j
j

Example
A Quantitative analyst has predicted the return of two stocks, A and B, under four conditions;
recession, stability, moderate grow and boom. The probability of the economy experience the
four conditions are respectively 0.1, 0.4, 0.3 and 0.2. Table 2.1 presents the predicted one
year return on a GHC 1000 investment in each stock under each economic condition.
Table 2.1

124
 Economic condition Stock A Stock B
Recession 40 -60
Stable economy 60 30
Moderate growth 120 350
Boom 200 400

Advise management based on;

i. Expected monetary value
ii. Expected opportunity loss
iii. Coefficient of variation
iv. Return - to- risk- ratio

Solution
n
i. Expected Monetary Value (mean / expectation) = X ij Pi
i 1

EMVA 40 0.1 60 0.4 120 0.3 200 0.2 GHC 104

EMVB 60 0.1 30 0.4 350 0.3 400 0.2 GHC 191
Stock B has the largest EMV and therefore should be selected
n
ii. Expected Opportunity Loss = LijPi
i 1

First find the optimum value from the stocks, thus comparing both stocks at each level of
economic conditions and selecting the highest value to be the optimum value.

From the table above, the optimum value at recession is 40, for stable economy is 60, for
moderate growth is 350 and for boom is 400.

Secondly, we have to find the difference between each stock level and the optimum level.

125
Economic condition A B Optimum Opportunity Loss (A) Opportunity Loss (B)
Opt Opt A Opt B

Recession 40 - 60 40 40 – 40 =0 40 – (-60) = 100

Stable economy 60 30 60 60 – 60 =0 60 – 30 = 30
Moderate growth 120 350 350 350 – 120 = 230 350 – 350 = 0

Boom 200 400 400 400 – 200 = 200 400 – 400 = 0

EOLA 0 0.1 0 0.4 230 0.3 200 0.2 GHC 109

EOLB 100 0.1 30 0.4 0 0.3 0 0.2 GHC 22

In conclusion, expected opportunity loss of stock B will be selected because it has the
smallest value of opportunity cost.

iii.

Coefficient of Variation CV 100%

Where Standard deviation is given by;

n
2
(Xi ) 2 P( X i )
i 1

2 2 2 2
A 40 104 0.1 60 104 0.4 120 104 0.3 200 104 0.2

409.6 774.4 76.8 1843.2 3104 55.71

2 2 2 2
B 60 191 0.1 30 191 0.4 350 191 0.3 400 191 0.2

6300.1 10368.4 7584.3 8736.2 32989 181.63

55.71
CV A 100% 100 % 53.57%
104

181.63
CVB 100% 100 % 95.09%
191

Stock A has the least CV and preferred.

126
 iv.

Return - to - risk ratio

For stock A ;
104
Return - to - risk ratio 1.87
55.71
For stock B ;
191
Return - to - risk ratio 1.05
181.63
Select Stock A because it has the largest RTRR

REVIEW QUESTIONS

1. An entrepreneur has a new project which he estimates will be completed and sold
within one year. The project will cost GH¢10,000,000 in total and since it is a new
project, he can only estimate the probability of various amounts of income from it.
Total Income (GH¢) 20,000,000 15,000,000 10,000,000 1,000,000
Probability 0.3 0.5 0.15 0.05
Should a bank manager lend him GH¢10,000,000 at 15% interest per annum to
finance the project, what is the entrepreneur’s expected profit?

2. An investor buys GH¢1,000,000 of shares with the object of making a capital gain
after one year. She believes that there is a 5% probability that the shares will double
in value, 25% that they will be worth GH¢1,500,000, a 30% probability that they will
only be worth GH¢500,000 and 40% probability that they will not change in value.
What is the expected monetary value of this investment, ignoring dealings cost?

3. A car accessory company ABC Ltd have developed a new car immobiliser and have
to decide whether to market the product nationwide, to sell by Mail Order or to sell
the patent to a large chain motor accessory shop. The estimated profits for each
decision depends on the state of the market which has been defined as high, medium,
and low. The probability that the state of the market will be either high medium or
low has been estimated as 0.25; 0.30 and 0.45 respectively. The expected profits (in
) is given in the table below.

127
 High Medium Low
Nationwide 95 52 (26)
Mail Order 48 24 19
Sell patent 25 25 25

What decision should the company take in order to maximise expected monetary
value?

4. An asset has the following possible returns with associated probabilities.

Possible returns 20% 18% 8% 0% -6%
Probability 0.1 0.45 0.30 0.05 0.10
Calculate the expected rate of return.

5. Securities and Y have the following characteristics.

Security Security Y
Return Probability Return Probability
30% 0.10 -20% 0.05
20% 0.20 10% 0.25
10% 0.40 20% 0.30
5% 0.20 30% 0.30
-10% 0.10 40% 0.10

Calculate the expected return for each security and comment on your results.

6. An automobile dealer stocks two models of a certain make of car. Model A costs
GH¢64,000,000 and Model B costs GH¢100,000,000. Of all customers expressing
interest in buying a car 6% buy model A, 4% buy model B, and the rest (90%) fail to
make a purchase. Find the expected earnings (sales) per customer.

128