Module 1

Topic 2 Probability

Learning objectives
In this topic you will be introduced to probability. Probability is the building
block of statistics. Much of the mathematical computation in statistics is based on
the principles of probability. After completing this topic, students should be able
to:
 Understand what is meant by the term probability.
 Understand the rules governing probability and how these rules help
formulate some of the rules of calculation.
 Understand some of the key terms used in probability.
 Distinguish between mutually exclusive and inferential collectively
exhaustive events.
 Be able to distinguish between dependent and independent events.
 Understand and apply Bayes theorem in solving problems.

In understanding the above, students will be able to use probability rules and
concepts to solve business problems. Although students might not be familiar
with some of the terms that we are about to use in this topic, at the end of this
topic, students will have an appreciation of the importance of these terms in
identifying and solving business-related problems.

Introduction
Before we can proceed to study statistics properly, we must first learn something
about probability. As decisions are made under the condition of uncertainty, we
use probability to measure uncertainty. Statistics, as a decision making tool, is
founded on probability theory.
Probability, as a language, is an old one and we know some of the principles of
probability intuitively.

Definition
In simple terms probability is the chance of an event happening. The probability
of an event E occurring is denoted by:
 0 P(E) 1
ie., the probability of an event can hold any value between 0 and 1 inclusive.
When P(E) = 1, then there is certainty of the event occurring and when P(E) = 0
there is no chance of that event occurring.

The personalistic theory

Sometimes probability is assigned to an event based on one's confidence,
knowledge, or intuition. Probability assigned in this fashion is said to be
personalistic, or subjective probability.
For example, a political commentator, based on his experience, might predict the
probability of the Labour Government being elected in the next federal election
as, say 60%.

The classical theory

Sometimes probability is calculated based on the items that make up the sample
space. When all points, or events, in a suitably constructed sample space are
equally probable, then the probability is calculated as below:

x No. of favourable outcome

P (E )  
n No. of total possible outcome

eg. The probability of drawing a king from a pack of cards:

4
P (king ) 
52
Probabilities assigned in this fashion are said to be objective probability.

Rules governing probability

1. Within a sample space S, if E is an event, then
0 P (E ) 1
that is, the probability of an event cannot be a negative or more than 1.
*Note: From now on, whenever you get a negative or a number largerthan
one, you are doing something wrong!
 Business Statistics I Guide 1  3

2. For any sample space S,

P (S ) 1
That is, the sum of all events in the sample space must equal 1.
An example of this can be illustrated through a deck of cards.
The probability of a King is 4/52, a queen is 4/52, a jack is 4/52, and so on.
If you were to add all the probabilities you should get 52/52 or 1.
3. With any sample space S, if A and B are two events, then
P(A  B) P(A )  P (B)  P(A  B)
This is called the General Additivity Rule.
P(A  B) is read as probability of A or B. The symbol U is called a
union sign.
P(A  B) is read as probability of A and B. The symbol  is called an
intersection sign.
4. Within any sample space S,
P(A ) 1  P (A )
P(A ) is read as the complement of A

Joint probability
A joint probability is the probability of more than one event occurring at the same
time. Mathematically, the joint probability of events A and B is given as:
P(A  B)
An example of a joint probability is the probability of a King of Diamonds being
drawn from a pack of playing cards. The card has to be a king as well as of
diamond suit. As there is only one card that matches this description in the pack
1
of cards, we assign the probability of 52 .

Mutually exclusive events

Events A and B are said to be mutually exclusive if there is no chance of the two
events occurring together. That is, when:
P(A  B) 0
An example of mutually exclusive events is drawing a Club and a red card from
one draw from a pack of playing cards. A club is a black card hence drawing a red
club is impossible!
The addition rule for mutually exclusive events therefore becomes:
P(A  B)  P(A)  P(B)
 Collectively exhaustive
Events are collectively exhaustive if they contain all possible outcomes of an
experiment or observation.
For example, drawing a black or a red suit from a pack of cards is collectively
exhaustive. A pack of cards contains only black suits or red suits. If you add the
cards with black or red suits, you will get 52 cards, which makes up the pack.
Therefore, P(Black U Red) = 1.


Activity
Example 1

If a die is tossed, what is the probability of

(i) scoring a three?
(ii) scoring at least a four?
What is the probability of drawing a 10 or a King from a standard deck of 52
playing cards?

*Hint: Remember the word ‘or’ should be substituted with the union sign!
Therefore you should be using the additivity formula.

continued over page

 Business Statistics I Guide 1  5

Activity continued

Example 2

(i) A card is drawn from a standard deck of cards. What is the probability of
getting a Jack or a Spade?
Here are some hints as to how to solve a probability problem:
First: Write out the question in probability notation. Note that the word
'or' suggests that you consider the additivity rule.
P(Jack or Spade) = P(Jack U Spade)
P(Jack U Spade) = P(Jack) + P(Spade) – P(Jack  Spade)
= 4/52 + 13/52 – 1/52
= 16/52
(ii) Consider a roulette wheel, which is divided into 37 compartments. Each
compartment bears a number, 0 (zero) and 1 to 36. Of these 37 numbers,
18 are black, 18 are red and the 0 is green. Determine the probability from
one spin of the wheel of obtaining:
(a) A red result
(b) A green result
(c) An odd result
(d) A number less than 15

Conditional probability
The probability of one event given that another event is known to have occurred
is called conditional probability. Sometimes the first event having occurred might
influence the probability of the second event occurring. Sometimes however it
doesn't.
Conditional probability is given the notation:
P(A |B) - This is read as
"probability of event A given B has occurred"
The formula for defining a conditional probability is:

P (A B )
P(A |B) 
P (B )
or
P (A B )
P(B| A ) 
P (A )

If we rearrange the above equation we get the multiplicative rule:

P(A  B) P (A |B ).P (B )
or
P(A  B) P (B | A ).P (A )

Dependent events
As mentioned earlier, sometimes the occurrence of the first event influences the
probability of the second event. If it does, then the events are called dependent
events ie. the outcome of one event is dependent upon the other. The above
formula is only used for events that are dependent.
An example of dependent event is drawing an ace given that an ace was drawn
and not replaced.

P(Ace) =

The probability of drawing an ace given an ace has just been drawn is:

P(Ace|Ace) = if the first ace is not replaced

Independent events
Two events are said to be independent if the occurrence of one event does not
alter the probability of the other event. An example of independent events is
throwing a die twice. The probability of getting a 5 from the first throw is the
same as from the second throw. That is :
1
P(5 from First throw) = 6
1
P(5 from second throw given you got 5 from first throw) = 6
 Business Statistics I Guide 1  7

Thus, the first throw did not alter the probability of the second. Notationally, an
independent event is shown as:
P(A |B) P ( A )
P(B| A ) P (B )
Therefore the multiplicative rule for independent events becomes

P(A  B) P (A ). P (B )


Activity

Example 3

(i) Give an example of an independent event.

(ii) The probability of an item being defective in a production process is 0.05.

What is the probability of picking two defective items in succession? (Items
being defective can be assumed to be independent)

continued over page

Activity continued

Example 4

An expert backgammon player is currently competing in two tournaments. He

estimates the probability of winning the first tournament as 0.6 and the
probability of winning the second tournament as 0.5. If he wins the first
tournament, he estimates the probability of winning the second tournament
increases to 0.65 because of an increase in confidence. What is the probability
of:
(i) the player winning both tournaments?
(ii) the player winning at least one tournament?
Before we attempt to solve for (i), it is advisable to assign a probability notation
to each of the numbers that you see in the question:
P(first) = 0.6 P(second) = 0.5 P(second | first) = 0.65
Note that the third sentence is a conditional statement. Whenever you see the
words ‘if’, ‘given’, the probability is a conditional one. The proper notation is:
P(second | first). This is read as ‘winning the second given the first is won’.
(i) the player winning both tournaments?
Note that the word ‘both’ implies winning the first and the second. You can see
that the word ‘and’ is implicit in the statement. Therefore, we need to use the
multiplicative rule.
P(A  B) = P(A|B) x P(B)
P(first  second) = P(second | first) x P(first)
P(first  second) = 0.65 x 0.6
= 0.39
(ii) the player winning at least one tournament?
Note that ‘at least one’ means one or the other. Thus the word ‘or’ is implied in
this statement. Therefore, we need to use the additivity principle.
P(A U B) = P(A) + P(B) – P(A  B)
P(first U second) = P(first) + P(second) – P(first  second)
= 0.6 + 0.5 – 0.39
= 0.71
*Important Note. In solving this problem, it is important that you first
write down in proper notation what each statement means. If you
practice doing probability problems in this fashion, you will be making
very few mistakes!
 Business Statistics I Guide 1  9

Contingency table
Sometimes information gathered in a survey can be expressed in a contingency
table. A contingency table is simply a cross-classification of data based on their
characteristics.
The probability on the margin of the tables is called marginal probability. A
marginal probability is the probability of a simple event occurring.
The probability within the table is called the joint probability. These probabilities
describe two or more events occurring. For example, the number 12 in the table
refers to students who are female and doing a postgraduate diploma.


Activity
Example 5

External Students enrolment in 1995 is cross-classified by three course levels

and their gender in the contingency table below:
Course level Male Female Total
Masters 40 31 71
Postgraduate Diploma 11 12 23
Graduate Diploma 118 311 429
Total 169 354 523
If a student is selected at random from this sample, what is the probability that:
(i) the student is enrolled in a postgraduate program?

(ii) the student is a male enrolled in a masters program?

(iii) the student is enrolled in the masters program and the graduate diploma
program?

(iv) If a student selected at random is a female student, what is the probability

that she is enrolled in a Masters program?

Is the level of the course the student enrols in dependent on their gender?

Note: See the back of this topic for a detailed explanation to this problem.
Bayes theorem
Bayes theorem is essentially another way of stating the definition of conditional
probability. It is a method of revising conditional probability based on new
information. It is essentially a process of determining posterior probabilities using
prior probabilities.
The use of Bayes formula can be quite tedious at times. That is why we will use
the decision tree diagram to help solve Bayes questions.
Bayes Formula:
P (A |Bi )P (Bi )
P (Bi | A ) 
P ( A |B1 )P (B1 )  P (A |B 2 )P (B2 ). .. P (A |B k )P (B k )
The principles of Bayes theorem can be illustrated easily using a tree diagram for
the following example.
Suppose you have two boxes labelled Box 1 and Box 2. In Box 1 there are 2 red
chalks and 4 blue chalks. In Box 2 there are 3 red chalks and 3 blue chalks.
The probability of picking any one of the two boxes is half. Within the boxes the
probability of picking a red or a blue chalk is given below:

BOX 1 BOX 2

2 - Red 3 - Red

4 - Blue 3 - Blue

P(Box 1) = 1/2 P(Box 2) = 1/2

P(Red|Box 1) = 2/6 P(Red|Box 2) = 3/6
P(Blue|Box 1) = 4/6 P(Blue|Box 2) = 3/6
Let's assume that the two boxes are covered and you are asked to pick a chalk.
Not knowing which box your hand is in, you pull out a red chalk.
What is the probability that the red chalk came from Box 1?
The notation for the question is P(Box 1|Red). This is the reverse of the
conditional probability that was initially provided. We can diagrammatically look
at the problem using a tree diagram.
 Business Statistics I Guide 1  11

The joint probability given in the right of the diagram above is the multiplication
of the marginal probability and the conditional probability. That is:
P(Red and Box 1) = P(Red|Box 1) x P(Box 1)
The total probability of drawing a red chalk is 5/12. That is, a red chalk could
have come from box 1 or box 2.
The probability of the red chalk coming from Box 1 is therefore

P (Box1 Red )
P (Box 1|Red ) 
P (Red )
2 2
12
P (Box 1|Red )  
5
12 5

Activity

Example 6

In the unit Business Statistics, past experience indicates that if students attend
all lectures and tutorials, they have an 85% chance of passing the course.
Students who do not attend all lectures and tutorials have a 30% chance of
passing the course. The lecturer estimates that 80% of students attend all
lectures and tutorials.
(i) Given that a student has passed the unit, what is the probability that he or
she attended all lectures and tutorials?
(Remember to first write down proper notation for the numbers in the
question)

(ii) What is the probability of a student failing the Business Statistics unit?
 Business Statistics I Guide 1  13

Formulas covered
1. Probability of an event, E,
x No. of favourable outcome
P (E )  
n No. of total possible outcome

2. Additivity rule for events that are not mutually exclusive

P(A  B) P(A )  P (B)  P(A  B)

3. Additivity rule for events that are mutually exclusive

P(A  B) P (A )  P (B )

4. Conditional probability
P (A B )
P(A |B) 
P (B )

5. Multiplicative rule for dependent events

P(A  B) P (A |B ).P (B )

7. Multiplicative rule for independent events

P(A  B) P (A ). P (B )

8. Bayes Theorem
Key terms
Simple Event: event that cannot be broken down into smaller
events.
Sample Space: the collection of all simple events.
Probability: the chance of an event happening.
Personalistic/Probability: where probability is calculated based Subjective on
one's confidence, knowledge, or intuition.
Classical/ Probability: where probability is calculated based Objective on
all possible outcomes (sample space).
Venn Diagram: a diagrammatic representation of probability of
events in a sample space.
Joint Probability: the probability of more than one event occurring at
the same time.
Mutually Exclusive: when there is no chance of two events occurring
together.
Collectively exhaustive: when events contain all possible outcomes of an
experiment or observation.
Dependent Events: when the occurrence of the first event influences the
probability of the second event.
Independent Events: when the occurrence of the first event does not
influence the probability of the second event.

Solution to examples
Example 1 – (i) P(three) = 1/6
(ii) P(  4) = 1/6 + 1/6 +1/6 = ½ (Note that at least 4 means –
greater than or equals to 4)
(iii) P(King U Ten) = P(King) + P(Ten) – P(King  Ten)
= 4/52 + 4/52 – 0 = 8/52

Example 2 – (i) solution is provided

(ii) (a) P(Red) = 18/37
(b) P(Green) = 1/37
(c) P(Odd) = 18/37
(d) P(<15) = 15/37 (Don’t forget to include zero)
 Business Statistics I Guide 1  15

Example 3- (i) You can use any example. How about tossing two coins
or throwing two dice. The probability of you passing the
statistics exam is independent of another external student
passing the exam.
(ii) P(Defective & = P (Defective) x P(Defective)
Defective)
= 0.05 x 0.05
= 0.0025
This is so because the events are independent.
Example 4 – solution provided
Example 5
This example is quite an easy one. Most of the answers are directly from the
table.
(i) P(Postgrad) = 23/523
This is a marginal probability that is read directly from the table.
The denominator should be 523.
(ii) P(Male  Masters) = 40/523
This is a joint probability. Again the denominator is 523. There are
40 male students doing the masters program from the total of 523
students.
(iii) P(Masters  Grad dip) = 0
Since these two category do not intersect the answer is 0. There
are no students doing the masters program and the graduate
diploma program.
(iv) P(Masters | Female) = 31/354
This is a conditional probability. The student picked is a female
student. If look at the female column, there are 354 of them. Out
of this 354 students, 31 are doing the masters program.
Alternatively we can use the conditional formula:

P(Masters | Female) = P(Masters  Female)/P(Female)

= 31/523/354/523
= 31/354
Are the events independent?
The appropriate proof is as follows:
If independent, then P(A|B) = P(A)
Ie. P(Masters | Female) = P(Masters)
31/354  71/523
Therefore we can conclude that the two events are not
independent.
A logical explanation to the above is that if the two events are independent then
the probability of a masters student given it is a female should be the same as the
probability of a masters student. If independent then, being a male or female
should not affect the probability of a student being a masters student.
Example 6
How did you find this example? It is important that we first write down the
notation for each of the numbers in the question:
P(Pass | Attend) = 85% P(Pass | Not attend) = 30%
P(Attend) = 80%
Once we have written down this, we should draw a tree diagram:

P(Pass) = 0.74
(i) P(Attend|Pass) = P(Attend  Pass)/P(Pass)
= 0.68/0.74
= 0.9189
(ii) P(Fail) = 1 – P(Pass)
= 1 – 0.74 = 0.26