STA110 (INTRODUCTION TO STATISTICS)

PART 3
Lecture Notes

◆ Introduction to Probability
◆ Use of Permutation and Combination in
Evaluating Probability
◆ Sampling from Tables of Random Numbers
◆ Types and Distribution of Random Variables

C. O. Odijie
Department of Statitsics
University of Benin
Benin City
Nigeria

© 2021 C. O. Odijie
.

These Lecture Notes are strictly for teaching purpose.

No part must be reproduced in any form
without the consent of the author.

© 2021 C. O. Odijie
All Rights Reserved.

Email: [email protected]
 Special Features
◆ The soft copy of this document is very flexible to navigate via links and
cross-references. Some are highlighted for quick reference. Feel free to click
any of them and return to your original position with keyboard shortcut
Alt+ i.e. Alternate + left arrow key if you are using a PDF reader that
supports the shortcut.
◆ There are copious examples on each topic and/or sub-topic.
◆ There are a good number of tutorial questions for practice under each topic
or after a group of sub-topics.
◆ There are special KEYNOTES to help summarise or emphasize important
points to note.
◆ Separated definitions, remarks, etc., which are numbered within sections e.g.
Remark 3.1 means “Remark 1 of section 3” for easy cross-referencing.

Literature
The student is encouraged to consult good textbooks and materials e.g. those by
Kroese and Chan (2014), Kroese (2018) and Blitzstein and Hwang (2019) among
others. See bibiography for details.
CONTENTS CONTENTS

Contents

Introduction to Probability 6
1 Random Experiment 6
1.1 Sample Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.2 Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.3 Events as Set Operations . . . . . . . . . . . . . . . . . . . . . . . . 9
1.4 Mutually Exclusive Events . . . . . . . . . . . . . . . . . . . . . . . 9
1.5 Mutually Exhaustive Events . . . . . . . . . . . . . . . . . . . . . . 9
1.6 Independent Events . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.7 Dependent Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Tutorial Questions Set 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2 Probability 12
2.1 Classical Definition of Probability . . . . . . . . . . . . . . . . . . . 12
2.2 Probability of Equally Likely Events . . . . . . . . . . . . . . . . . 12
2.3 Addition Law of Probability . . . . . . . . . . . . . . . . . . . . . . 15
2.4 Multiplication Law of Probability . . . . . . . . . . . . . . . . . . . 15
2.5 Conditional Probability . . . . . . . . . . . . . . . . . . . . . . . . . 16
Tutorial Questions Set 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Use of Permutation and Combination in Evaluating Prob-

ability 23
3 A Brief Review of Permutation and Combination 23

4 Application of Permutation and Combination to Probability 25

Tutorial Questions Set 3 29

Sampling from Tables of Random Numbers 30

5 Table of Random Numbers 30
5.1 How to read a table of random numbers . . . . . . . . . . . . . . . 30
5.2 Importance and Limitations . . . . . . . . . . . . . . . . . . . . . . 31
Tutorial Questions Set 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 4

CONTENTS CONTENTS

Types and Distribution of Random Variables 33

6 Random Variable 34
6.1 Types of random variables . . . . . . . . . . . . . . . . . . . . . . . 34
6.2 Distribution of Random Variables . . . . . . . . . . . . . . . . . . . 34
6.3 Normal Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Tutorial Questions Set 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
6.4 Exponential Distribution . . . . . . . . . . . . . . . . . . . . . . . . 41
Tutorial Questions Set 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
6.5 Binomial Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Tutorial Questions Set 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
6.6 Poisson Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Tutorial Questions Set 8 . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
6.7 Hypergeometric Distribution . . . . . . . . . . . . . . . . . . . . . . 52
Tutorial Questions Set 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Bibliography 56

Appendix 57

Answers to Selected Tutorial Questions 57

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 5

Random Experiment

Introduction to Probability
1 Random Experiment
A basic notion in probability theory is that of a random experiment. A ran-
dom experiment is one whose outcomes cannot be predetermined (i.e. out-
comes cannot be determined in advance). In other words, the outcomes depend on
chance/probability. Although the outcomes cannot be predetermeined, they can
still be analyzed because of the knowledge of their general nature.
An outcome is the result of a single trial of a random experiment. A very basic
example of a random experiment is the flip of a coin. It is a random experiment
because the outcome (a head appearing or a tail appearing) cannot be known
before the flip, even though it is general known that each outcome would always
be one of these two. You may have seen this random experiment several times
being carried out before the start of a football match. Other examples of random
experiments include:

Examples Set 1
(a) counting the number of cars that pass through UNIBEN Main Gate
between 8 a.m. and 4 p.m. in a day,
(b) randomly selecting 50 Statistics students in 100 Level in a particular
session and measuring their heights,
(c) measuring the weight of 100 randomly selected yam tubers harvested by a
farmer,
(d) tossing a fair die,
(e) measuring the amount of rainfall in Benin City in May.

The goal of probability theory is to describe these random experiments with ap-
propriate mathematical models with which they can be analyzed. The models
are made up of three building blocks: a sample space, a set of events, and a
probability. Let us examine each of these building blocks one after the other.

1.1 Sample Space

Although we have said that each outcome of a random experiment cannot be
predicted with certainty, we can however specify a set of possible outcomes for a
random experiment. This is usually the first constituent of our model for a random

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 6

Random Experiment 1.2 Events

experiment. For example, although we cannot tell which face of a fair die would
show in a single toss, we know that it will definitely be any one of the numbers 1,
2, 3, 4, 5 or 6. This set of possible outcomes is what we refer to as sample space.
Definition 1.1 (Sample Space) The sample space Ω of a random experiment
is the set of all possible outcomes of the experiment.

Examples Set 2
The following are some examples of random experiments and their sample spaces.
1. Throw two fair dice one after the other,
Ω  tp1, 1q, p1, 2q, . . . , p1, 6q, p2, 1q, p2, 2q, . . . , p2, 6q, p3, 1q, p3, 2q . . . , p3, 6q, . . . ,
p6, 1q, p6, 2q, . . . , p6, 6qu.
2. The number of arriving calls at a customer care centre during a specifi
ed time interval,
Ω  t0, 1, 2, 3, . . . u, the set of non-negative integers.
3. The lifetime of an electric bulb (measured in days),
Ω  R , the set of positive real numbers.
4. Flip two fair coins together,
Ω  tHH, HT, T H, T T u.
5. Measure the height (in centimeters) of 10 randomly selected students,
Ω  tpx1 , x2 , . . . , x10 q : xi ¡ 0, i  1, 2, . . . , 10u.

Remark 1.1 In the last example, px1 , x2 , . . . , x10 q represents the outcome that the
height of the first selected student is x1 , the height of the second selected student is
x2 , and so on, where each student’s height must be positive (denoted by xi ¡ 0, i 
1, 2, . . . , 10). It is wiser to use the largest possible sample space when the limits
are not certain. This is the case of example 3. Of course, we know that an electric
bulb will not have a lifetime that spans the positive real axis (i.e. last forever) but
it is better to assign the whole set of positive real numbers since we are not certain
what the maximum life span of an electric bulb would be. A similar explanation
goes for the heights of students which may not even exceed 300cm (3 meters) in
example number 5.

1.2 Events
When the outcomes of a random experiment are considered as well-defined subsets
relative to the sample space, they are called events.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 7

Random Experiment 1.2 Events

Definition 1.2 (Event) An event is a well-defined subset of the sample space.

Although events are usually groups of outcomes as subsets of the sample space,
sometimes single outcomes as well make up events (simple events, e.g. the out-
come/event of getting a six in a single throw of a die). Events are usually rep-
resented with block letters such as A, B, C, and so on. An event A occurs if the
outcome of the experiment is at least one of the elements in A.

Examples Set 3
Let us consider some events in the random experiment of throwing two fair dice
given in the first example of Examples Set 2 of sub-section 1.1. First, the sample
space is given as
$ ,
'
'p q
1, 1 , p1, 2q, p1, 3q, p1, 4q, p1, 5q, p1, 6q,/
/
'
'
'
'p q
2, 1 , p2, 2q, p2, 3q, p2, 4q, p2, 5q, p2, 6q,/
/
/
/
Ω
&
p3, 1q, p3, 2q, p3, 3q, p3, 4q, p3, 5q, p3, 6q,.
'
' p4, 1q, p4, 2q, p4, 3q, p4, 4q, p4, 5q, p4, 6q,/
/
'
'
'
' p5, 1q, p5, 2q, p5, 3q, p5, 4q, p5, 5q, p5, 6q,/
/
/
/
%
p6, 1q, p6, 2q, p6, 3q, p6, 4q, p6, 5q, p6, 6q -
We define the following events.
1. The event A of obtaining a sum of face numbers equal to 12 is given as
A  tp6, 6qu.
2. The event B that the second die is 2 is given as
B  tp1, 2q, p2, 2q, p3, 2q, p4, 2q, p5, 2q, p6, 2qu.
3. The event C that both dice show equal face numbers is given as
C  tp1, 1q, p2, 2q, p3, 3q, p4, 4q, p5, 5q, p6, 6qu
4. The event D that a sum of face numbers greater than 12 appears is the
empty set tu or ϕ. That is, an impossible event in this case.
5. The event E that a sum of face numbers less than or equal to 12 appears is
the sample space Ω itself.

Some events (called compound events) combine more than one simple events.
Examples of compound events defined on the same sample space Ω in Examples
Set 3 are given in the next set of examples.

Examples Set 4
1. The event F of obtaining a sum of face numbers equal to 12 or a sum of face
numbers equal to 8 is given as
F  tp2, 6q, p3, 5q, p4, 4q, p5, 3q, p6, 2q, p6, 6qu

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 8

Random Experiment 1.3 Events as Set Operations

2. The event G that 2 appears on the second die or 1 appears on the first die
is given as
G  tp1, 2q, p2, 2q, p3, 2q, p4, 2q, p5, 2q, p6, 2q, p1, 1q, p1, 3q, p1, 4q, p1, 5q, p1, 6qu.
3. The event H that the first die shows an odd number and the second die
shows an even number, and is given as
H  tp1, 2q, p1, 4q, p1, 6q, p3, 2q, p3, 4q, p3, 6q, p5, 2q, p5, 4q, 5, 6qu.

Remark 1.2 Note that the events in Examples Set 3 are all simple events unlike
those in Examples Set 4 which are compound events. The event G for instance is a
union of the separate simple events “the event that 2 appears on the first die” and
“the event that 1 appears on the second die”. We can use set operations (mainly
union and intersection) to describe compound events as will be seen later. At this
point, the student is assumed to have a basic knowledge of set notations.

1.3 Events as Set Operations

Events are sets and so we can apply the usual set operations to describe them as
follows:
◆ the set A Y B (A union B) is the event that A or B or both occur,
◆ the set A X B (A intersection B) is the event that A and B both occur,
◆ the event Ac or A1 (A complement) is the event that A does not occur,
◆ if A € B (A is a subset of B) then event A is said to imply event B.

1.4 Mutually Exclusive Events

Definition 1.3 (Mutually Exclusive Events) Two events A and B which have
no outcomes in common, that is, A X B  ϕ, are called disjoint events or mu-
tually exclusive events.
Mutually exclusive events cannot occur together. E.g. the event of getting a head
and the event of getting a tail in the flip of a fair coin are mutually exclusive. We
cannot get a head and a tail together as an outcome in a single flip.

1.5 Mutually Exhaustive Events

Definition 1.4 (Mutually Exhaustive Events) Two events A, B € Ω are said
to be mutually exhaustive if A Y B  Ω. That is, if their union “exhausts” the
sample space.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 9

Random Experiment 1.6 Independent Events

A typical pair of events that are both mutually exclusive and exhaustive is any
given event and its compliment (e.g. A and A1 ).

1.6 Independent Events

Definition 1.5 (Independent Events) Two events are said to be independent
if the incidence or occurrence of one does not affect the probability of incidence
or occurrence of the other.
Events arising from the random experiment of throwing a fair coin or a fair die are
independent events as the outcome of one throw does not affect the probability of
the outcome of another. Also, events form random experiments of drawing items
with replacement are independent events.

1.7 Dependent Events

Definition 1.6 Two events are said to be dependent if the occurrence of one
affects or changes the probability of occurrence of the other.
Events from experiments of drawing objects without replacement are dependent,
since the occurrence of one event immediately affects the probability of occurrence
of the other(s).

Figure 1: Venn diagram depicting three events as sets.

It is useful to represent events as sets in Venn diagrams as depicted in Figure 1. In

the figure, the events A and C as well as the events B and C are disjoint (mutually
exclusive) events. However the events A and B have a common intersection, and
hence not mutually exclusive. We now list more deductions from the Venn diagram
(Figure 1) in set notations as follows:

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 10

Random Experiment Tutorial Questions Set 1

KEYNOTES 1

1. A X B  ϕ but A X C  ϕ and B X C  ϕ,
2. pA X B 1q X pA1 X B q  ϕ i.e. pA X B 1q and pA1 X B q are disjoint events, 
3. A  pA X B 1 q Y pA X B q & B  pA1 X B q Y pA X B q ,
4. A Y B  A Y pA1 X B q  B Y pA X B 1 q.

These deductions will be very useful in the proof of some properties of probability
later.

Tutorial Questions Set 1

1. Write down the following events in set notation.
(a) The event that the sum of the faces that appear when two dice are
thrown is 10 or more.
(b) The event that a machine lives less than 1000 days.
(c) The event that the height of a student among ten randomly selected
students is not more than 200cm (2m).
(d) The event of getting at least two heads in the throw of 3 fair coins.
2. A die is cast and a coin is flipped. Write down
(a) the sample space for the random experiment,
(b) the event that two tails appear,
(c) The event that a six appears.
(d) Let A be the event that an odd-numbered face appears on the die and
let B be the event that a prime-numbered face appears on the die.
Then, find (i) A Y B, (ii) A X B, (iii) A1 , (iv) B 1 , (v) pA X B qc , (vi)
pA Y B qc.
(e) Interpret each of the events in 2 (d) (i) –(vi)
3. Represent the following events as set operations in Venn diagrams.
(a) A Y B, (b) pA Y B q1 , (c) A X B X C, (d) A Y B Y C, (e) A1 .
4. Write down De Morgan’s laws for a collection of sets Ai , i  1, 2, . . . , n.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 11

Probability

2 Probability
A very important tool in the model for a random experiment is the specification
of the probability of occurence for the events in the sample space of an experi-
ment.This probability tells us how likely it is that a particular event will occur.
Definition 2.1 (Probability) A probability P is a rule or function which as-
signs a non-negative real number to each outcome/event of the sample space Ω of
a random experiment, and which satisfies the following axioms.
Axiom 1. PpAq ¥ 0 for an event A € Ω.
Axiom 2. PpΩq  1.
Axiom 3. For any sequence A1 , A2 , A3 , . . . of mutually disjoint events in Ω, i.e.
Ai X Aj  ϕ, for i  j, we have that

8
¤ 8̧
P Ai  PpAi q

i 1 
i 1

This definition is sometimes refered to as axiomatic definition of probability.

2.1 Classical Definition of Probability

A ‘simpler’ definition of probability, often called classical definition sees the prob-
ability of an event A as the number of outcomes in A divided by the total number
of possible outcomes. That is,

npAq
PpAq 
npΩq
where npAq means the number of elements (single outcomes) in A and npΩq means
the number of elements in the sample space, Ω.

2.2 Probability of Equally Likely Events

When each outcome of a random experiment always has equal probability of oc-
curring as the other outcomes, we say that they are equally likely . For example,
the two sides of an unbiased (fair) coin are equally likely to appear in a flip, each
having probability 0.5 of appearing. Similarly, each of the six sides of a fair die
is equally likely to appear in a throw, each having probability 1/6 = 0.1667 of
appearing. On the other hand, some outcomes have different probabilities of oc-
curring. Such outcomes are not equally likely. For instance, if a jar contains 6 red,
5 green and 8 blue balls and a single ball is picked at random from the jar. Then,

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 12

Probability 2.2 Probability of Equally Likely Events

(i) The probability of choosing a red ball is given by

Ppred ballq  
no. of red balls 6
total no. of balls in the jar 19
(ii) The probability of choosing a green ball is given by
Ppgreen ballq  
no. of green balls 5
total no. of balls in the jar 19
(iii) The probability of choosing a blue ball is given by
Ppblue ballq  
no. of blue balls 8
total no. of balls in the jar 19
Clearly, these outcomes are not equally likely. Notice that the probabilities in-
creased, the higher the number of the coloured balls. Here, a blue ball is more
likely to be picked than a red or a green ball because it has the largest number
in the jar, and hence, a larger probability 8/19 i.e. 8 out of 19 chances of being
picked. We can thus intuitively imply a first property of probability that:

If A and B are events in any sample space Ω and npAq ¥ npB q then PpAq ¥ PpB q.
npAq
The proof is directly from classical definition of probability. That is
npΩq
¥ nnppBΩqq
since in the numerators, npAq ¥ npB q for a fixed denominator npΩq

Theorem 2.1 (Properties of Probability) For any non-empty events

A, B € Ω, we have that:

1. Ppϕq  0, ϕ is the empty set.

2. A€B ñ PpAq ¤ PpB q.

3. 0 ¤ PpAq ¤ 1.

4. PpA1 q  1
PpAq.

5. PpA Y B q  PpAq PpB q
PpA X B q.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 13

Probability 2.2 Probability of Equally Likely Events

Proof.

Property 1. From set theory, Ω  Ω Y ϕ (read up idempotent laws of set

operation) where Ω and ϕ are disjoint. Thus, from axiom 3, we have that PpΩq 
PpΩq Ppϕq, but by axiom 2, PpΩq  1 and so we have Ppϕq  0
Property 2. Consider Figure 2.

Figure 2: Venn diagram depicting A € B.

Clearly, B  A YpA1 X B q where A and pA1 X B q are disjoint events. Thus applying
axiom 3, we have
PpB q  PpAq PpA1 X B q, and hence PpAq ¤ PpB q follows, since PpA1 X B q ¥ 0.

Property 3. Follows from axioms 1 and 2.

Property 4. Now, Ω  A Y A1 (students should verify this with the aid of venn
diagram, or otherwise, and also verify that A and A1 are mutually exclusive and
exhaustive with respect to the sample space Ω. See definitions 1.3 and 1.4). Since
A X A1  ϕ, applying axiom 3 yields PpΩq  PpAq PpA1 q, and we have that
PpA1 q  1
PpAq because PpΩq  1 by axiom 2.

Property 5. Recall deductions 3 and 4 of KEYNOTES 1. Specifically,

B  pA1 X B q Y pA X B q (2.1)

and
AYB  A Y p A1 X B q (2.2)

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 14

Probability 2.3 Addition Law of Probability

Now, since A and pA1 X B q are disjoint, we can apply axiom 3 to equation (2.2) to
obtain:

PpA Y B q  PpAq PpA1 X B q (2.3)

Next, since pA1 X B q and pA X B q are disjoint events, we apply axiom 3 to equation
(2.1) to obtain
PpB q  PpA1 X B q PpA X B q,
ñ
PpA1 X B q  PpB q
PpA X B q (2.4)

Finally, substituting PpA1 X B q from equation (2.4) into equation (2.3), we have
PpA Y B q  PpAq PpB q
PpA X B q■

2.3 Addition Law of Probability

This law/rule applies to the union of two or more events. For two events A, B € Ω,
it is the probability that either A or B or both occur. It states that

PpA or B q  PpAq PpB q
PpA and B q (2.5)

This can be written in set operation notation as

PpA Y B q  PpAq PpB q
PpA X B q (2.6)

When the two events are mutually exclusive, that is, pA X B q  ϕ implying
that PpA X B q  Ppϕq  0, the addition law reduces to

PpA Y B q  PpAq Pp B q (2.7)

Addition law can be extended to three events as follows

PpA Y B Y C q  PpAq PpB q PpC q
PpA X B q
PpA X C q
PpB X C q PpA X B X C q

(2.8)
The addition law becomes difficult to write out clearly for four or more events.

2.4 Multiplication Law of Probability

This law or rule relates to finding the probability of two or more events happening
together i.e. at the same time or trial. For two events A and B in a given sample
space Ω, it states that:

PpA and B q  PpAq  PpB |Aq (2.9)

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 15

Probability 2.5 Conditional Probability

where PpB |Aq is called the conditional probability that event B occurs given
that event A has occurred and is defined in definition 2.2.
Note that PpA and B q  PpA X B q as may be seen in the literature. Note also that
A X B  B X A by commutative property of intersection.
When events A and B are independent, the Multiplication Law reduces to

PpA and B q  PpAq  PpB q

Or alternatively,
PpA X B q  PpAq  PpB q (2.10)

2.5 Conditional Probability

When probabilty of an event is considered relative to an event that has already
occurred we are dealing with conditional probabilty.
Definition 2.2 (Conditional Probability) Let A and B be two non-empty events
defined on the the sample space, Ω. Then, the probability that event B occurs given
that event A has occurred is called the conditional probability of B given A and is
defined by
PpA X B q
PpB |Aq  , PpAq ¡ 0
PpAq
(2.11)

It immediately follows from equation (2.10) that if A and B are independent,

PpB |Aq  PpB q (2.12)

Similarly,
PpA|B q  PpAq (2.13)

Examples Set 5
A. Addition Law (With Mutually- and Non-Mutually Exclusive Events)
1. You draw one card from a standard deck of cards, find the probability that
it is (a) either an ace or a diamond? (b) either a king or a queen (c) either a
card numbered 5 or a heart (d) odd-numbered card or a joker. (See appendix
for a pictoral sample of a deck of cards)

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 16

Probability 2.5 Conditional Probability

(a) Let A be ace and D be diamond

PpA or Dq  PpAq PpDq
PpA and Dqq
but PpAq  4{52 and PpDq  13{52.
Here the events are not mutually exclusive because the card can be an
ace and a diamond (i.e. ace of diamonds) with probability
PpA X Dq  1{52 (because there is only one ace of diamonds).
Thus, PpA or Dq  4{52 13{52
1{52  4{13
(b) Let K be king and Q be queen. Then,
PpK or Qq  PpK q PpQq
PpK and Qq
A card cannot be a king and a queen. Therefore,
PpK or Qq  4{52 4{52
0  2{13
(c) Let F be a card numbered five and H be heart.
PpF or H q  PpF q PpH q
PpF and H q
A card can be numbered 5 and a heart (i.e. five of hearts). Thus,
PpF or H q  4{52 13{52
1{52  4{13
(d) Let O represent odd-numbered card and J represent joker.
PpO or J q  PpOq PpJ q
PpO and J q
A card cannot be odd-numbered and a joker because a joker has no
number but the letter J.
There are 4 odd-numbered cards (i.e. cards with the number 3, 5, 7
and 9: see Figure 4 in appendix) in each of the four suits. Hence, there
are 16 such cards. So,
PpO or J q  16{52 4{52
0  5{13
2. Given that PpAq  0.6, PpB q  0.4, PpA X B q  0.24 and
PpC q  0.6, PpDq  0.3, PpC X Dq  0.22
(a) Find PpA Y B q
(b) Are the events A and B mutually exclusive? Why?
(c) Are A and B independent? Give reason(s).

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 17

Probability 2.5 Conditional Probability

(d) Are A and B mutually exhaustive? Why?

(e) - (h): attempt 2 (a) - (c) above, replacing A, B with C, D respectively.

Now,
(a) PpA Y B q  PpAq PpB q
PpA X B q  0.6 0.4
0.24  0.76.
(b) The events A and B are not mutually exclusive. This is because their
probability of occuring together is not zero. i.e. PpA X B q  0.24  0
(c) A and B are independent. This is because PpA X B q  PpAq  PpB q i.e.
0.24 = (0.6)(0.4)
(d) A and B are not mutually exhaustive because PpA Y B q  0.76 1 
PpΩq. Stated in words, the probability of their union is 0.76, which is
less than 1, which is the probability of the sample space Ω. It means
their union is not equal to or does not exhaust the sample space.
(e) - (h) are left as exercise for the students to tackle following (a) - (d).
3. In a class of 50 students, 27 are boys and 23 are girls. On a particular test,
14 boys and 15 girls made an A grade. If a student is chosen at random from
the class, what is the probability of choosing a girl or a student who made
an A grade?
Let G be the event of picking a girl and A be the event of picking a student
who made an A grade.Then,

PpG or Aq  PpGq PpAq
P pG and Aq

 23{50 29{50
15{50
 37{50.

B. Multiplication Law (With Dependent or Independent Events)

4. Suppose you draw two cards at random from a standard pack of cards one
after another, without replacing the first card. What is probability that the
first card is the queen of hearts, and the second card is a heart?
Let X be the event that the first card is the queen of hearts and Y be the
event that the second card is a heart. The two events are dependent events
because the first card is not replaced.
There is only one queen of hearts in a deck of 52 cards. So,

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 18

Probability 2.5 Conditional Probability

PpX q  1{52.
Now,
PpX and Y q  PpX q  PpY |X q (multiplication rule or law)
After drawing the first card being queen of hearts, there are now 12 hearts
left in the hearts suit and 51 cards left in the pack. Hence, the probability
that the second card is a heart given that the first card is the queen of hearts
is given by
PpY |X q  12{51,
so that

PpX and Y q  p1{52q  p12{51q

 1{221
 0.0045.
This is an example of ordered drawing without replacement. We can see that
the denominator, 52  51  52P 2 , while the numerator is 12  12P 1 as will be
detailed in sections 3 and 4.
5. An unbiased coin is tossed once and a fair die is rolled once. Find the
probability of landing a head on the coin and rolling a 5 on the die.
The outcome on the coin (head, H or tail, T) is clearly independent of the out-
come on the die (any face number in the set of face numbers t1, 2, 3, 4, 5, 6u.
Hence,

PpH on the coin and 5 on the dieq  PpH on the coinq  Pp5 on the dieq
 p1{2q  p1{6q
 1{12 or 0.0833.
6. A card is chosen at random from a deck of 52 cards. It is then replaced and
a second card is chosen at random. What is the probability of choosing a
joker and then a seven?
Now, the two events are independent because the first card was replaced
before choosing the second. So,

Ppjoker and sevenq  Ppjokerq  Ppsevenq

 p4{52q  p4{52q
 2{13 or 0.1538.
C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 19
Probability 2.5 Conditional Probability

C. Conditional Probability
7. A Statistics lecturer gave his class tests on two courses. 25% of a the class
passed both tests and 42% of the class passed the first test. What percent
of those who passed the first test also passed the second test?
A little care has to be taken here not to make the mistake of evaluating
the required probability as under multpiplication law. Let’s get out the
probabilities of the various events to see a clear picture of the problem.
Let the events that the first and second tests were passed be A and B,
respectively. Then, from the problem,
PpA X B q  0.25 (% that passed both tests)
PpAq  0.42 (% that passed first test)
Clearly now, those who passed the first test, who also passed the second test
ñ those who passed the second test given/knowing that they have passed
the first test and is given by definition as
PpB |Aq  PpA X B q{PpAq
Substituting values, we have:
PpB |Aq  p0.25q{p0.42q  0.5952.
That is about 59.52%  60% passed the second test given that they passed
the first test.
8. The probability that it is Friday and a student is absent is 0.05. Assume
a typical 5 school days per week. What is the probability that a student
selected from attendance register at random is absent given that the day is
Friday? Interprete your result in terms of % chance.
Let F be the event that a day choosen at random is a Friday and A be the
event that a student is absent. Then
PpF q  1{5  0.2 (one of five days)
PpA X F q  0.05
The required probability is

PpA|F q  PpA X F q{PpF q

 p0.05q{p0.2q
 0.25.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 20

Probability Tutorial Questions Set 2

There is 25% chance that a student selected randomly is absent given that
the day is Friday.
9. If the probability that Jane buys a new pair of shoes given that she travels
abroad is 0.75. What is the probability that Jane travels abroad and buys a
new pair of shoes if the probability that she travels abroad is 0.25.
Let B be the event that Jane buys a new pair of shoes and T be the event
that Jane travels abroad.
Given probablilities are
PpB |T q  0.75, PpT q  0.25
Required probability is PpB X T q
Using the definition of conditional probability, we have
PpB |T q  PpB X T q{PpT q
That is
PpB X T q
0.75 
0.25
6 PpB X T q  p0.75qp0.25q  0.1875.

Tutorial Questions Set 2

(For problems 1 – 15, visit: https://www.mathgoodies.com/lessons/vol6)
1. On New Year’s Eve, the probability of a person having a car accident is 0.09.
The probability of a person driving while intoxicated is 0.32 and probability
of a person having a car accident while intoxicated is 0.15. What is the
probability of a person driving while intoxicated or having a car accident?
2. A day of the week is chosen at random. What is the probability of choosing
a Monday or Tuesday?
3. In a pet store, there are 6 puppies, 9 kittens, 4 gerbils and 7 parakeets. If
a pet is chosen at random, what is the probability of choosing a puppy or a
parakeet?
4. A number from 1 to 10 is chosen at random. What is the probability of
choosing a 5 or an even number?

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 21

Probability Tutorial Questions Set 2

5. A single 6-sided die is rolled. What is the probability of rolling a number

greater than 3 or an even number?
6. The probability of a New York teenager owning a skateboard is 0.37, of own-
ing a bicycle is 0.81 and of owning both is 0.36. If a New York teenager is
chosen at random, what is the probability that the teenager owns a skate-
board or a bicycle?
7. Spin a spinner numbered 1 to 7, and toss a coin. What is the probability of
getting an odd number on the spinner and a tail on the coin?
8. A jar contains 6 red balls, 3 green balls, 5 white balls and 7 yellow balls.
Two balls are chosen from the jar, with replacement. What is the probability
that both balls chosen are green?
9. Four cards are chosen from a standard deck of 52 playing cards with replace-
ment. What is the probability of choosing 4 hearts in a row?
10. A nationwide survey showed that 65% of all children in the United States
dislike eating vegetables. If 4 children are chosen at random, what is the
probability that all 4 dislike eating vegetables? (Round your answer to the
nearest percent.)
11. Two cards are chosen at random from a deck of 52 cards without replacement.
What is the probability of choosing two kings?
12. Two cards are chosen at random from a deck of 52 cards without replacement.
What is the probability that the first card is a jack and the second card is a
ten?
13. On a math test, 5 out of 20 students got an A. If three students are chosen
at random without replacement, what is the probability that all three got
an A on the test?
14. Three cards are chosen at random from a deck of 52 cards without replace-
ment. What is the probability of choosing an ace, a king, and a queen in
order?
15. A school survey found that 7 out of 30 students walk to school. If four
students are selected at random without replacement, what is the probability
that all four walk to school?
Others: Algebraic and Numerical Evaluations, Proofs
16. If two events A and B are mutually exclusive, then PpA|B q  ?

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 22

A Brief Review of Permutation and Combination

17. Show that PpA Y B q  PpAq PpB q
PpAq PpB q if A and B are independent
and
PpA Y B q  PpAq PpB q
PpAq  PpB |Aq if A and B are dependent.



18. Show that P pA Y B q|A  P pA Y B q|B  1
19. Given that X and Y are independent events and PpY |X q  0.25, Pp X q 
0.65, Find PpX Y Y q
20. Given that PpS and T q  .4, PpT |S q  .5, find PpS q
21. *Two cards are drawn from a deck of 52 cards one after the other without
replacement. What is the probability that one is a spade and the other is a
numbered red suit card? (Hint: 1. Notice that there is no particular order
in which the two events must occur. 2. Study Figure 4 properly to get the
events correctly).

Use of Permutation and Combination in

Evaluating Probability
3 A Brief Review of Permutation and Combination
The problem of counting usually arises in many real life situations, especially that
of arrangements and selection. For example, how can we form a committee of five
persons in a group of ten persons where everyone is eligible to be a part of the
committee? In how many ways can this be done? The problem becomes even
more difficult to handle when there are some restrictions on how the committee,
for example, is to be formed. Problems like this one are usually dealt with under
permutation and combination, under the broad topic called combinatorics.
When we want to determine the number of possible arrangements in a set for
which the order of arrangements matters, we are dealing with permutation. For
example, the letters ABC, ACB, BAC, BCA, CAB and CBA are the six (6) pos-
sible arrangements (permutations) of the individual letters A, B and C. Suppose
we are to arrange only two of the three letters at a time. Then we would have
the following possible arrangements: AB, BA, AC, CA, BC, CB. Clearly, we can
see that the order of appearance of the letters in the arrangement matters. Hence,
AB is not the same as BA in permutation.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 23

A Brief Review of Permutation and Combination

On the other hand, when the order of arrangement does not matter in a selection
from a set, we are dealing with combination. Considering the same letters as before,
there is only one way of combining the letters A, B and C, that is tA, B, C u. As
long as each of the letters appears in the selection or combination, the position
(order) in which a letter appears does not matter in the combination. Suppose
again that we are to take only two letters at a time. Then we would have AB, AC,
and BC. Obviously the reverse ordering arrangements (BA, CA, and CB) are not
necessary here. Mathematically stated,



KEYNOTES 3

n
1. A k-combination from a set of n items is written as or nC k and
k
defined as 
n
k
 k!pnn!
kq!
2. A k-arrangement from a set of n items is written as nP k and defined
as
Pk 
pn
kq!  npn
1qpn
2q . . . pn
k 1q
n n!

where ! is the factorial symbol and

n!  npn
1qpn
2qpn
3q    3  2  1

So far, the objects being arranged have been distinct. If some of the set of objects
to be permuted have identical constituents, the permutation procedure will be
different as in the following example.

Example 3.1
In how many ways can the letters of the word BEER be arranged?
The word BEER has four letters and two are identical. Suppose we label the two
identical E’s as E1 and E2 . All the letters would be distinct and there will be
4
P 4  4!  24 ways of arranging the word. But since the two E’s are actually
identical, the number of permutations will be reduced because we cannot identify
which of the E’s is used where. E.g. the arrangements BE1 RE2 and BE2 RE1 , if
the E’s were labeled, would just be BERE as the E’s are actually identical. So
will every other arrangement where the two identical letters alternate positions be
reduced to just one. That is, the two E’s would be arranged in 2! ways if they were

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 24

Application of Permutation and Combination to Probability

distintint. Therefore, we divide the total number of permutations obtained

if all items were distintct by the factorial (!) of the number of identical
items. The correct number of arrangements in our example is
4!
2!
 12 ways. The
same procedure applies when there are more than one set of identical objects like
in the word SUCCESS which has
7!
3!2!
 420 permutations. The division by 3!2!
is as a result of letter S appearing three times and letter C appearing twice.

Other forms of permutation with special restrictions like a particular letter must
be in a specific position can be carefully handled with a combination of the rules
learned so far. The student is advised to read up more about permutations and
combinations as this is onlya brief review.

Summarily,

1. Ordered drawing/selection of items without replacement ñ Permutation

2. Unordered drawing/selection of items without replacement ñ Combination
KEYNOTES 4

3. The relationship between nP k and nC k is given by

(a) nP k  pnC k q  k!

or
n
(b) nC k 
Pk
k!
4. Total arrangements of n items where p items are identicals of one kind, q
items are identicals of another kind,    
n!
p!q!   
5. Cyclic permutation of n distinct items is given by pn
1q!.

Note: Summary point number 3a shows how permutation contains combina-

tion: select k distinct items first (combination) and arrange into all possible
k! ways of ordering them.

4 Application of Permutation and Combination to Probability

Many random experiments such as random drawing of cards from a deck of cards,
random picking of marbles from a box, random sitting arrangements, etc., rely

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 25

 Application of Permutation and Combination to Probability

on permutation and combination to know the number of ways that outcomes can
possibly be obtained. Then probabilities can be evaluated as ratios of possible
outcomes to total outcomes. Permutation and combination often relate to random
experiments where drawing or selection is done without replacement. Let’s see
an example to explain this.

Example 4.1
Consider a problem of drawing three marbles from an urn containing four marbles
without replacement. In how many ways can this been done?

We can approach the problem as follows:

Step 1 Write out three boxes lll
Step 2 There are four marbles and so the first box has four ways of being filled with
a marble from the urn. 4 ll
Step 3 Without replacement, the next box has 3 ways of being filled with any of the
3 remaining marbles in the urn. 4 3 l
Step 4 The final box has 2 ways of being filled. 4 3 2
Now, 4  3  2  4P 3 
p4
3q!  24.
4!

So, there are 24 ways of drawing drawing 3 marbles from an urn containing 4
different marbles.
Suppose we replace each marble selected at random before selecting the next one.
There would always be four marbles to pick from and the result of selecting three
marbles from the urn would be 4  4  4  64. That is there would be 43  64
ways of selecting 3 marbles from an urn containing 4 marbles if replacement is
allowed after each random selection.

NOTE: Ordered vectors such as p4, 3, 2q, p3, 2, 4q, p2, 3, 4q, . . . , are usually used to
represent permutations of a single set of combination t4, 2, 3u, say. Here, p2, 3, 4q 
p4, 3, 2q  p3, 2, 4q  . . . being vectors but t3, 2, 4u  t4, 3, 2u  t2, 3, 4u  . . . as
equal sets denoting a single combination where the order of arrangement of the
elements does not matter.

Examples Set 6
1. What is the probability that in 5 tosses of a fair coin we get exactly 3 Heads?

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 26

Application of Permutation and Combination to Probability

2. Draw 10 cards at random from a deck of cards. What is the probability that
we draw 3 Hearts and 5 Spades?
3. What is the probability that all 20 people in a group have different birthdays?
(Assume 365 days per year)
4. What is the probability of choosing an ATM PIN of 4 distinct digits? Assume
that every possible arrangement of four digits is eligible as an ATM PIN.
5. Find the probability that a word chosen at random from all the possible
permutations of the word AV AT AR starts with letter T

Solution
In all the the problems, let Ω denote the sample space and let A be the required
event so that npΩq is the total number of outcomes and npAq is the number of
outcome in event A.

1. Tossing a fair coin five times gives npΩq  25  32 (a coin has two sides i.e.
two possible outcomes, then five tosses)

The event A of getting exactly 3 Heads can be obtained in
5
3
 10 ways.
That is, the combination tH, H, H, T,T u of 5 outcomes where 3 outcomes
must be H can be arranged in
5!
3!2!
 5
3
since we have a total of 5 letters
out of which 3 H’s are identical and 2 T ’s are identical. (Refer to the last
bullet item of KEYNOTES 4)

6 PpAq  10
32
 0.3125

We shall revisit this problem in Example 6.4 under binomial distribtuion in

sub-section 6.5.

2. 10 cards are drawn at random from 52 cards ñ npΩq 
52
10
The event A of getting 3 hearts and 5 spades from a suit of 13 hearts and a
suit of 13 spades, respectively can be done in
    
npAq 
13 13 26 26
where the starred combination is as a result
3 5 2 2
of the remaining 2 cards since 10 cards were drawn. That is, 3 hearts, 5
spades and 2 other cards from the remaining 52
p13 13q  26 cards

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 27

Application of Permutation and Combination to Probability

  
13 13 26

6 PpAq  3

5
52
2
 0.0076
10
3. The probability that out of 20 people in a group, all have different birthdays
(based on 365 days in a year) is
365
P 20
36520
 0.5886. How?
First, taking 365 days in a year, the first person can have any of the days as
birthday. Next, for the second person to not share the first person’s birthday,
we must “remove” the first person’s birthday from the days of the year.The
the second person now has 364 days left. This reasoning carries through to
the 20th person who now has 346 days left to have a birhtday if it must not
“clash” with the previous 19 birthdays and we have:
365  364  363  362    346
Pp20 people have different birthdaysq 
365  365  365  365    365
365
The numerator is clearly P 20 while the denominator is 36520
4. npΩq  104 . One way of seeing this clearly is the usual
(number of outcomes for each trial of the experiment)number of trials .
That is (10 digits to pick from)4 slots or trials .
Another way is to list all the possible outcomes (PINs) and see how many
are there. That is, how many numbers are there from 0000, 0001, 0003, . . . ,
9999? The answer is 9999
0000 1  10, 000  104 numbers because any
sequence of consecutive natural numbers in ascending order a1 , a2 , . . . , an has
pan
a1 1q terms. e.g. 4, 5, 6, 7, 8 has 8
4 1  5 terms which can easily
be verified here by counting them.
For the required event A, there are 10 digits from which 4 distinct digits
can be selected in this problem and the order of the digits matters, i.e.
select four distinct digits and then arrange into all possible 4! ordered forms.
For instance (4,1,3,5) will be a different PIN from (1,4,5,3), both having four
distinct digits as required. So the outcomes in event A involves a permutation
problem. See again summary “Note” of KEYNOTES 4.
So, npAq  10P 4
 5040
6 PpAq  10000
5040
 0.5040

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 28

Tutorial Questions Set 3

5. Number of ways of arranging the word AV AT AR is npΩq   120

6!
(3!,
3!
because letter A appears 3 times).
The event A that a word chosen at random from these 120 permuted words
will begin with letter T can be done in npAq  20 ways as follows:
There are 6 letters to permute. So draw six slots like llllll
The first box must be letter T and so it has only one way of being filled. 1
lllll
The remaining 5 boxes can then be filled with a letter successively in 5  4 
3  2  1  5!  120 ways divided by 3! because letter A appears 3 times.
That is 120{6  20 ways.
So, PpAq  20{120  1{6  0.1667

Tutorial Questions Set 3

1. 8 cards are picked at random from a deck of 52 cards. What is the probability
that they are all from the same suit? What does your answer say about the
event “they are all from the same suit”?
2. An urn contains 6 blue balls and 4 green balls. 3 balls are picked at random
one after the other without replacement. What is the probability that (a)they
are all blue (b) they are all green (c) all of the same colour (d) only one is
green (e) two are blue
3. What is the probability that in a group of 10 people, there is at least one
shared birthday?
4. What is the probability that a hacker guesses a 4-digit ATM PIN correctly?
Assume that all 4-digit numbers including 0000 qualify as ATM PIN.
5. What is the probability that a 5-digit password ends in digit 5 if no digit
must be repeated in the password.
6. Find the probability that a word chosen randomly from all the permutations
of the word BU T T ER has the two T 1 s together.
7. Evaluate the probability of obtaining 8 heads and 2 tails in 10 flips of a fair
coin.
8. Show that (a) nP 1  nC 1  n (b) nC k  nC n
k (c) nC n  1 (d) nP n  n!
C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 29
Table of Random Numbers

Sampling from Tables of Random Numbers

5 Table of Random Numbers
This is a table that contains a series of digits (0 to 9) randomly arranged in
rows and columns, usually in groups of 5 digits. It is used when performing
random experiments and random sampling for researches. There is no relationship
whatsoever between any two digits or two series of digits, so that one cannot guess
any set of digits by knowing the set before or after it. Different tables contain
different series of random digits. A sample copy is reproduced in Table 1.

Table 1: A sample of table of random numbers

(Source: https://mathbitsnotebook.com/Algebra2/Statistics/STrandomtable.html)

5.1 How to read a table of random numbers

Depending on the user and the purpose of use, there are several ways to use a
table of random numbers. One of such ways is highlighted as follows:

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 30

 Table of Random Numbers 5.2 Importance and Limitations

Suppose we are to simulate 5 scores from an exam marked over 100%.

Step 1 Choose a starting point randomly.
This can easily be done by pointing to the table without looking. Suppose
we land on the group of digits 89116 in the table above i.e. 1st row, 3rd
entry.
Step 2 Choose whether to read along rows (horizontally) or along columns (verti-
cally) starting from that entry.
Suppose we choose to read along row in this example.
Step 3 Select the number of digits as appropriate.
Since we are sampling for exam scores marked over 100, we must pick three
digits at a time to get any of 000 - 100. Whether to pick only the first three
digits of each 5-digit group along the row or to pick the immediate next three
digits after the current one also depends on the researcher. However, for the
purpose of uniformity in assessment, we shall specify which one to follow. In
our example here, let us specify that we pick only first three digits of each
5-digit group along each row. In our case of landing on 89116 (1st row, 3rd
entry), we should have picked 891 but this is a greater than 100. Ignore this,
and continue row-wise. Hence, from the next group 05542, we pick 055 as
first sample (since 000 ¤ 055 ¤ 100 i.e. it lies within the required range of
samples). Continuing in this way, the next correct pick is 087 (from 08747
in row 2, entry 6), and so on.
Our final result will be: 55, 87, 70, 72, 12
Remark 5.1 It is usually easier and faster to sample from every digit along each
row or column from the starting point that was randomly chosen in case of personal
research. What would be the result if we followed this process?

5.2 Importance and Limitations

The use of random number table will help the researcher to obtain independent
random samples. For fairly large samples, usually n ¡ 30, the samples generated
should exhibit properties of normality which is a very desirable property of random
samples. Samples with these properties are easier to analyze in practice.
However, the process of sampling with random number tables can be very slow
and tedious for large experiments. Computer software packages have now replaced
tables of random numbers in simulating samples in large experiments. Examples

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 31

Table of Random Numbers 5.2 Importance and Limitations

of such computer packages include Statistical packages like Minitab, SPSS, R, and
others such as MATLAB, Mathematica, etc.

Examples Set 7
1. Simulate a random experiment of flipping a coin 10 times. Start from
column 1, entry 2 of Table 1 and consider only the first digit of each group
along the columns.
2. Simulate an experiment of rolling a fair die 10 times. Start selection from
row 3, group entry 4 and sample row-wise, considering all digits in each
group.

Solution
1. Let 1 represent a head and let 0 represent a tail. We are required to start
from column 1, entry 2. That is the entry group 46132 in Table 1 and we are
to consider only the first digit of each column. Reading downwards and then
from up to down the next column (column 2), etc, (up to the 4th column in
this example) we have the result as follows:
1, 0, 0, 1, 0, 1, 0, 0, 0, 0 = H, T, T, H, T, H, T, T, T, T
2. We are to start sampling from row 3 and entry 4, sampling along rows and
considering all digits (not only first digits of groups).
Now, the 3rd row from 4th entry reads: 97402 27569 90184 . . .
We are picking any of the numbers 1, 2, 3, 4, 5, or 6 to represent the face
numbers of a fair die till we pick the 10th eligible number from all the digits
of each 5-digit group.
Our result is: 4, 2, 2, 5, 6, 1, 4, 2, 3, 3
This second example buttresses Remark 5.1.

Assignment
1. A teacher wishes to to measure the heights of 8 pupils randomly selected
from his class. Design and perform an experiment, with the aid of a random
number table, to select the pupils to partake given that there are 30 pupils
in the class.
2. Perform the experiment in question 1 of Examples Set 7 ten times, using
different starting points and sampling direction of your choice. Take record

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 32

Table of Random Numbers Tutorial Questions Set 4

of the number of heads xi in the ith experiment. Then find the mean of the
number of heads in the i  10 experiments. Is the mean close to 0.5?
If you can perform the experiment a large number of times, say N ¥ 100,
1 ¸
N
your mean number of heads xi will be  0.5 for sure (i.e. the coin
N i1
turns up heads about np  n = half of the total number of times the coin
1
2
was thrown in each experiment in the long run of N experiments). This is a
practical demonstration of the expectation of a binomial random variable X
(representing the number of heads in the throw of a fair coin some number
of times, n, in this case).

Tutorial Questions Set 4

1. A researcher wishes to select a random sample of 18 students from a class of
50 students. Suppose the students are numbered 1 - 50. Generate a random
sample using Table 1. Start from row 5, entry 5 and sample along rows,
considering all the digits in a group.
2. Simulate a random sample of ten results of a “Yes” or “No” interview ques-
tion responses. Start from column 1, entry 7 of Table 1, sample along columns
taking only the first digit or first set of digits (whichever is appropriate for
the problem) of each group. State clearly the assumption you made to rep-
resent a “Yes” or a “No” response.
3. With the use of a random numbers table, explain how you would carry out
an experiment to sample 20 out of 200 workers in a company needed for an
interview.
4. Obtain a random sample of 10 2-digits numbers between 50 and 99 (both
inclusive), starting from the first entry of column 2 of Table 1. Consider
every digit along columns.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 33

Random Variable

Types and Distribution of Random

Variables
6 Random Variable
A random variable is a quantity whose values depend on outcomes of random
experiment or phenomenon. More formally, a random variable is a function defined
on a probability space that assigns real values to the outcomes of a sample space. A
random variable is denoted with a capital letter while a particular value it assumes
is written in small letter.

6.1 Types of random variables

There are two types of random variables:
1. discrete random variable
2. continuous random variable
Discrete random variables take any of a specified finite countable list of values.
Continuous random variables take any numerical value in an interval or collection
of uncountable intervals.

6.2 Distribution of Random Variables

Random variables are described by probability distributions. While the distribu-
tion of a discrete random variable is called its probability mass function (PMF),
the probability distribution of a continuous random variable is called probability
density function (PDF).
Some examples of discrete probability distributions are:
◆ binomial distribution
◆ Poisson distribution
◆ hypergeometric distribution
◆ geometric distribution
◆ negative binomial distribution
Examples of continuous probability distributions include, among many others,

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 34

Random Variable 6.3 Normal Distribution

◆ normal distribution
◆ exponential distribution
◆ Student’s t distribution
◆ chi-square distribution

In this course, we shall introduce the binomial, Poisson, hypergeometric, normal

and exponential distributions.

6.3 Normal Distribution

The normal distribution, also called Gaussian distribution, is one of the most
widely used distributions in Statistics. It is a continuous distribution with a bell-
shaped density curve. Its applicability is wide because many other distributions
tend to converge to the normal distribution via the so-called Central Limit Theo-
rem (CLT) as the sample size increases. So the normal distribution can approxi-
mate many other distributions via CLT. This will be discussed in details in higher
level courses in Statistics.
A continuous random variable X which is normally distributed has the following
PDF
# +
1
x
µ
f pxq  ?1 exp

8 8
2
, x (6.1)
σ 2π 2 σ

The parameters of the distribution are

µ  mean (also median and mode) of the distribution;
8 µ 8

σ 2  the variance of the distribution; σ 2 ¡ 0

6.3.1 Some properties of normal distribution

1. Mean = median = mode =µ
2. Data are evenly spread around the mean, 50% above and 50% below the
mean. That is, it is symmetrical about the mean = median =mode
3. The curve is bell-shaped
4. Variance = σ 2 ¡0
C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 35
Random Variable 6.3 Normal Distribution

A random variable X which is normally distributed with mean µ and variance σ 2 ,

with PDF in equation 6.1 is described for short as

X ∽ N pµ, σ 2 q

The cumulative distribution function(CDF) of any continuous random variable X

evaluates the probability that X lies in the interval p
8, xs for a given value x of
the random variable X and is defined by

»x
PpX ¤ xq  F pxq  f ptqdt (6.2)

8
So, for X ∽ N pµ, σ 2 q in our case, the CDF is given by

» x
# +
PpX ¤ xq  F pxq  ?1
2
σ µ
1
t 2
exp dt (6.3)
σ 2π

8
It is not easy to integrate equation 6.3 to compute probabalities for values of X in
a given interval. So, the normal random variable X is usually standardized using
the following transformation

Z  X σ
µ (6.4)

where the random variable Z, distributed as

Z ∽ N p0, 1q

is called the standard normal random variable with mean µ  0 and variance
σ 2  1. Its probability distribution function is obtained by substituting σ  1 and
the tranformation 6.4 into the PDF in equation 6.1 and is given by

f pz q  ?
12 z2
8 8
1
exp , z (6.5)
2π

The CDF of the the standard normal random variable Z is thus given by

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 36

Random Variable 6.3 Normal Distribution

» z


Φpz q  ?1 exp
21 t2 dt (6.6)
2π

8
The CDF of Z in equation 6.6 is clearly simplier than that of X in 6.3. Infact,
cumulative probabilities PpZ ¤ z q  Φpz q (see figure 3) for different values of z
have been computed and tabulated in standard normal tables popularly called the
Z-score table or Z table.
Given a problem under normal distribution therefore, it is better to tranform to
X
µ
standard normal distribtuion via Z  before solving, since we cannot solve
σ
the the normal CDF algebraically or analytically.

Figure 3: Area under standard normal curve from
8 to z

6.3.2 Some Properties of Standard Normal Distribution

1. PpZ ¤ zq  Φpzq
2. Ppa ¤ Z ¤ bq  Φpbq
Φpaq
3. PpZ ¥ z q  1
PpZ z q
4. Φp
aq  1
Φpaq
5. Φp0q  0.5 i.e. z  0, “splits” the distribution into two “symmetric halves”

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 37

Random Variable 6.3 Normal Distribution

6. EpZ q  µ  0 and VpZ q  σ 2  1.

Table 2: An example of how to read a standard normal table in a “reverse” manner

6.3.3 How to read a standard normal table for z-value given Φpz q
Suppose we are to find the z-value up to which area under the standard normal
curve from
8 is Φpz q  0.9307. Since 0.9307 ¡ 0.5, we search under the positive
Z-scores for the given probability, 0.9307. We can only find a value close enough
in this example and that is 0.9306. So we trace to the left of the table and to the
top to get 1.4 and 0.08 respectively (see Table 2). We add these two values to get
the required z-value, i.e. 1.48. This value is sometimes written as Φ
1 pz q. Hence
Φ
1 p0.9307q  1.48.

Examples Set 8
1. Find Pp1 ¤ X ¤ 4q given that X ∽ N p2, 25q.
2. Given that X ∽ N p6, σ 2 q and PpX ¡ 16q  0.0228, find σ 2 of X.
3. The scores in a test are estimated to be randomly distributed with mean 55
and variance 36. If the cut-off mark if set at 60, what percentage of
students who took the test scored below the cut-off mark?
4. Find k such that Pp|Z | ¤ k q  0.95 where Z ∽ N p0, 1q
5. Find µ and σ 2 given that PpX ¤ 2q  0.5 and PpX ¥
2q  0.7881 for a
normal random variable X.
Solution

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 38

Random Variable 6.3 Normal Distribution

1. µ  2, σ 2  25 ñ σ  5, Z  X σ
µ
Hence,

¤ 4q  P 1
5 2 ¤ Z ¤ 4
5 2


Pp1 ¤ X
 Pp
0.2 ¤ Z ¤ 0.4q
 Φp0.4q
Φ p
0.2q 
 Φp0.4q
1
Φp0.2q (property 4 applied)
 Φp0.4q Φp0.2q
1
 0.6554 0.5793
1
6 Pp1 ¤ X ¤ 4q  0.2347

2. µ  6, σ2 ? Pp X ¡ 16q  0.0228, Z  X σ
µ .
Thus,

¡ 16 σ
6  0.0228


Pp X ¡ 16q  P Z


1
P Z  0.0228
10
σ

10
1
Φ  0.0228
σ

10
Φ
σ
 0.9772

Reading from the table, the z-value up to which the cumulative

probabilities from the left tail equals 0.9772 is 2. That is, 0.9772  Φp2q or
Φ
1 p0.9772q  2

10
6 Φ σ  Φp2q
ñ 10σ
 2 and σ  5
So, σ 2  25
3. Let X be the scores from the test.
µ  55, σ 2  36 ñ σ  6, Z 
X
µ
60
55
σ

Required probability is PpX 60q  P Z  Φp5{6q  0.7977
6
So, about 79.77%  80% scored below cut-off mark 60.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 39

Random Variable Tutorial Questions Set 5

4. Pp|Z | ¤ k q  0.95 ñ Pp
k ¤ Z ¤ k q  0.95 as given.

Now, Pp
k ¤ Z ¤ k q  Φpk q
Φp
k q  2Φpk q
1  0.95
Hence, Φpk q   0.975
1.95
2
From the table, 0.975 area under the Z curve from the left tail corresponds
to the z-value 1.96.
Hence, k  1.96

¤ 2q  P Z ¤ 2
σ µ  0.5 and


5. PpX

PpX ¥
2q  1
PpZ
2q  1
P Z

2
µ  0.7881
σ

2
µ

2
µ
That is, Φ
σ
 0.5 and Φ σ  1
0.7881

2
µ

2
µ
ñΦ σ
 0.5 and Φ σ
 0.2119
From tables, Φp0q  0.5 and Φp
0.8q  0.2119

ñ Φ 2
σ µ  Φp0q and Φ
2σ
µ  Φp
0.8q

6 2
µ  0 and
2
µ 
0.8
σ σ
ñ µ  2 and σ  5
So, mean µ  2 and variance σ 2  25

Tutorial Questions Set 5

1. If X is a normally distributed random variable with mean 30 and standard
deviation 4. Find
(a) PpX 40q
(b) PpX ¡ 21q
(c) Pp30 X 35q
2. The speed limit on a certain road is 100 kmph. Speed of cars on that road
follows a normal distribution with mean µ  90 kmph and variance σ 2 
100 kmph. If a car is selected at random on the that road,
(a) find the probability that it travels above the speed limit.
(b) find the probability that it adheres to the speed limit.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 40

Random Variable 6.4 Exponential Distribution

(c) Find the probability that a car chosen at random on the road travels
with a speed between the mean speed and the speed limit.
3. The ages Y of pupils in a class are normally distributed as Y ∽ N p10, 4q.
Find the probability that the age of a pupil selected at random
(a) exceeds the mean age of pupils in the class
(b) lies within 2σ of the mean age.
(c) is above 12 years.
4. Suppose the weight of male adults follows a normal distribution with mean
weight 70 Kg and variance 49. Find the probability that a male adult selected
at random
(a) weighs at least 60 Kg
(b) weighs less than 60 Kg
(c) weighs between 90 Kg and 98 Kg

6.4 Exponential Distribution

The exponential distribution is a continuous probability distribution which de-
scribes a Poisson process in which events occur in a continuous and independent
manner at a constant average rate. It is a special case of the gamma distribution.
A continuous random variable (r.v.) X is said to follow the exponential distribtu-
tion and written for short as X ∽ Exppλq, if it has the PDF given by
#
λe
λx , if x ¥ 0
f px, λq  (6.7)
0, otherwise

where λ ¡ 0 is the parameter of the distribution called the rate parameter.

The distribution is mainly used to describe time between events such as time
between calls in a telephone service center, time between customers’ arrival at a
bank, time between equipment failure or breakdown, etc.
Some properties of exponential distribution includes

1. Mean EpX q 
1
λ

2. Variance VpX q 
1
λ2

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 41

Random Variable 6.4 Exponential Distribution

The CDF of a random variable X which follows the exponential distribtuion is

given by
»8
F pxq  PpX ¤ xq  λe
λtdt
0 (6.8)
 1
e
λx, x¥0

Example 6.1 Given that jobs arrive at a business center at the rate of 2 jobs
per minute on the average, what is the probability that the center waits less than
30 seconds for the next job? What is the maximum waiting time between two job
arrivals with 95% confidence?
λ  2 (rate of job arrival is 2 jobs per minute on the average). Waiting for 30
seconds  waiting for 0.5 minutes.
Let T be the wating time between two jobs. The PDF of T is given by
f pt; λq  2e
2t , t ¥ 0
and then, the required probabilty is

» 0.5
PpT ¤ 0.5q  2e
2tdt
0
 1
e
2p0.5q
 1
e
1
 0.6321
Let t* be the maximum wait time between two jobs with 95% confidence. Then
probability that the waiting time T does not exceed t* with 95% confidence is
given by
Pp T¤ tq  0.95
1
e
2t  0.95


e
2t  0.05


2t  lnp0.05q
lnp0.05q
t 

2
 1.5 minutes
This means that the business center is 95% confident that it will not wait more
than 1 minute 30 seconds between any two jobs.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 42

Random Variable 6.4 Exponential Distribution

Example 6.2 The lifetime T (years) of a device follows the exponential

distribution given by
f ptq  e
t , t ¥ 0

Find,
(a) The expected lifetime of the device
(b) the lifetime L which a typical device is 70% certain to exceed
(c) If five devices are to a dealer, find the probability that at least one of them
will have a lifetime less than L.

Solution
f ptq  e
t , t ¥ 0 implies that λ  1. Hence,
(a) Expected value EpT q  1{λ  1
(b) We are to find L such that PpT ¡ Lq  0.7
That is,

1
PpT ¤ Lq  0.7
1
p1
e
L q  0.7
e
L  0.7
6 L 
lnp0.7q
 0.3567 years
This means that with 70% confidence, the least lifetime of a typical device
is 0.3567 years.
(c) Assuming that the lifetime of each device is independent of another’s, then
Ppa device’s lifetime is not Lq  PpT ¥ Lq  0.7
and by independence,

Ppall five devices’ lifetime are not Lq  p0.7q5

 Ppnone of five devices’ lifetime is Lq.

Now,
Ppat least one of five devices’ lifetime is Lq
 1
Ppnone of five devices’ lifetime is Lq

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 43

Random Variable Tutorial Questions Set 6

 1
p0.7q5
 0.8319

Tutorial Questions Set 6

1. Suppose that the time (days) between births in a particular hospital follows
the exponential distribution with rate 0.2. Find (a) the mean time between
births in the hospital. (b) The variance of time between births in the hospital
(c) What is the probability that there is a birth in the hospital in the next
(i) 5 days? (ii) 10 hours? (iii) 20 minutes?
2. Suppose that the time between emergency calls in a fire station follows the
exponential distribution with an average rate of 2 calls per day.What is the
probability that the firemen are not called in 3 days?
3. The time (minutes) between customers arrival at a bank follows the expo-
nential distribution, Expp0.75q. What is the probability that (a) 3 minutes
elapse before the next customer arrives (b) no new customer arrives in the
next 50 seconds (c) a customer arrives in less than 30 seconds after the pre-
vious customer.
4. Find the mean and variance of the exponential random variable whose aver-
age rate of occurence is 0.625.

6.5 Binomial Distribution

The binomial distribution is a discrete probability distribution of the number of
successes in a sequence of independent trials of a Bernoulli experiment (an ex-
periment where there are only two outcomes such as head or tail in the flip of a
coin, yes or no to a question, male or female as gender, and so on). The binomial
distribution has two parameters: n, which is the number of Bernoulli trials and p,
which is the probability of success in each trial.

6.5.1 Assumptions of binomial distribution

The basic assumptions of the binomial distribution include the following.
1. There are two mutually exlusive outcomes in each trial of the experiment
2. The probability of success remains the same (constant) from trial to trial
3. The trials are independent

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 44

Random Variable 6.5 Binomial Distribution

4. The number of experiment trials is fixed.

The probability mass function (PMF) of a discrete random variable which follows
the binomial distribution is given by

PpX  kq  p p1
pqn
k , k P t0, 1, 2, 3, . . . nu
n k
k (6.9)
0¤p¤1

where
n
k
 pn
n!kq!k!  npn
1qpn
2k!q    pn
k 1q
is a binomial coeffi-
cient.
Alternatively, substituting q  1
p into equation 6.9, we have

n k n
k
PpX  kq  p q , k P t0, 1, 2, 3, . . . nu
k
0¤p¤1
(6.10)

q  1
p or p q 1
We write X ∽ B pn, pq to mean that the random variable X follows the binomial
distribution with parameters n and p.
The mean of a binomial variable X is given by EpX q  np and the variance,
VpX q  npp1
pq or npq where q  1
p

Example 6.3
According to a consumer survey in a village 15% of young adults (18 -24 years of
age) own a mobile phone with internet access . In a random sample of 30 young
adults from the village, let X be a number who own a mobile phone with
internet access. If X follows a binomial distribution,
(a) what is the value of p?
(b) What is the probability that exactly half own a mobile phone with internet
access?
(c) What is the probability that they all have a mobile phone with internet
access?

Solution
(a) Let X be the number of young adults chosen at random who have a mobile
phone with internet access. Clearly, either a young adult selected at

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 45

Random Variable 6.5 Binomial Distribution

random has a mobile phone with internet access or does not and 15%
belong to those who have. Therefore the probability of success p  0.15.
(b) Half of 30 young adults selected is 15.

Pp X  15q  30 p0.15q15p0.85q15
15
 5.93  10
6
(b) All of them.

PpX  30q  30 p0.15q30p0.85q0
30
 1.92  10
25

Example 6.4
Let us revisit problem 1 of Examples Set 6.
The probability of obtaining exactly 3 Heads in 5 tosses of a fair coin can be
evaluated with the binomial distribution where p  1{2 per trial and n  5 trials,
so that given H as the random variable representing number of heads in 5 tosses
of a fair coin, we have

PpH  3q  p0.5q3p0.5q2
5
3

 53 p0.5q5
 0.3125
which is the same as the result  in solution of problem 1 of Examples Set 6

Example 6.5
In a survey with questionnaires, 25% of the respondents answered a YES to a par-
ticular question. 5 questionnaires are picked at random. What is the probability
that (a) at least 3 of the respondents answered a YES to the question (b) exactly
2 answered a YES to the question (c) none answered a YES to the question (d)
all five respondents answered a YES to the question (e) at least one respondent
answered a YES to the question.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 46

Random Variable 6.5 Binomial Distribution

Let X denote the number of respondents chosen at random who answered a YES
to the question. Now,

p  0.25 ñ q  1
p  0.75, n  5, PpX  kq  p p1
pqn
k
n k
k

(a) At least 3 (i.e. 3 or more).

PpX ¥ 3q  PpX  3q PpX  4q PpX  5q 

 53 p0.25q3p0.75q2 54 p0.25q2p0.75q3 5
5
p0.25qp0.75q4
 0.0879 0.0146 0.0010
 0.1035
(b) Exactly 2.

PpX  2q  52 p0.25q2p0.75q3
 0.2637
(c) None.

PpX  0q  50 p0.25q0p0.75q5
 0.2373
(d) All.

PpX  5q  55 p0.25q5p0.75q0
 0.0010
(e) At least one: PpX ¥ 1q
To avoid evaluating 5 combinations of probabilities, we shall use the reverse
probability:
PpX ¥ 1q  1
PpX 1q

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 47

Random Variable Tutorial Questions Set 7

Since x P t0, 1, 2, 3, 4, 5u, X 1 means X  0. So

PpX ¥ 1q  1
PpX  0q

 1
50 p0.25q0p0.75q5
 1
0.2373
 0.7627
Remark 6.1 Care must be strictly taken when evaluating probability of discrete
random variables that lie in a discrete interval. Unlike the continuous case where
strict inequalities do not alter the results of integration because the interval is
continuous, the result will be altered in discrete case. For example PpX ¥ 1q 
PpX ¡ 1q whereas the two inequalities will give the same result in continuous
distribution cases. This is so because in dicrete intervals there are no elements
between any two integers whereas there are infinite number of elements between
two integers in a continuous interval.
Remark 6.2 Some experiments may not inherently be Bernoulli in nature but
the definition of the event of interest can guarantee the application of binomial
distribution. For instance, the experiment of randomly asking people their choice
of institution is not Bernoulli in nature (there are more than two possible an-
swers/outcomes). However, the random variable X that a particular institution
e.g. UNIBEN is a choice in consideration as the random variable of interest can
be treated as a binomial variable when we group all other choices of institution
together as failure, provided the probability p of success (the respondent having
UNIBEN as their choice) is known for a fixed number of trials n.
Remark 6.3 The words “success” and “failure” used in probability sense do not
have their usual meaning in other real life usage. Success could be the breakdown
of a machine within its specified life span if that is the event of interest or
under consideration and “failure” could mean that the machine keeps working
after this specified life span has expired.

Tutorial Questions Set 7

1. A couple wish to have four children. What is the probability that exactly
two are boys? Assume that the couple have equal chances of giving birth to
a girl or a boy.
2. An unbiased coin is flipped 20 times. What is the probability of getting

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 48

Random Variable 6.6 Poisson Distribution

(a) exactly 10 heads?

(b) at most 5 heads?
3. There are 5 black balls and 5 white balls in a box. 4 balls are drawn randomly
one after the other with replacement. Find the probability that
(a) *all are black or all are white [Hint: “or” stands for union i.e. addition
law and let the random varaible X represent the number of black (or
white) balls in 4 draws]
(b) 3 are black

6.6 Poisson Distribution

The Poisson distribution is a discrete probability distribution which measures the
probability of events occurring in a specified time interval or space interval. The
events are independent and the average rate λ of occurrence of events is constant
over the interval of time or space and independent. It a good distribution to
measure events with rare occurrences over a long interval of time or space.The
Poisson distribution approximates the binomial distribution when the probability
of success p in each trial of the binomial distribution is small and the number of
trials n is very large. That is, when p Ñ 0 and n Ñ 8 the mean of the binomial
distribution np Ñ λ, the rate parameter of the Poisson distribution. The Poisson
distribution seems to be the discrete counterpart of the exponential distribution
where both have events occuring in Poisson process at a constant average rate
λ. However, while exponential distribution measures time between events in a
continuous interval, the Poisson distribution measures the number of occurrences
of an event in a specified discrete interval of time or space. Both distributions
are defined on infinite intervals (one continuous and the discrete other discrete, of
course).
A discrete random variable X follows the Poisson distribution with parameter
λ ¡ 0 if the PMF of X is given by
k
λ
PpX  kq  λ k!e , k  0, 1, 2, . . . (6.11)

where
e  2.71828 . . . is the Euler’s number
k is the number of occurrences
k!  k pk
1qpk
2q    2  1 is the factorial of k.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 49

Random Variable 6.6 Poisson Distribution

Interestingly, the mean and variance of the Poisson distribution are equal
and the same as the rate paramater λ. That is
EpX q  VpX q  λ, for a Poisson random variable, X.
In this topic we shall write X ∽ P oispλq to mean X follows a Poission distribution
with mean λ.

Example 6.6 Given that a random variable X follows a Poisson distribution

with mean 3.5, find PpX  6q.
Now, λ  3.5.
6
3.5
Pp X  6q  3.5 6!e
 0.0771

Example 6.7 The number of accidents per week in a particular district is known
to follow a Poisson distribution with rate parameter 0.5. Find the probability
that
(a) in a particular week there will be:
(i) less than 2 accidents,
(ii) more than 2 accidents,
(b) in a three week period there will be no accidents
Let Y be the number of accidents in one week. Then Y ∽ P oisp0.5q with PMF
given by

k
0.5
PpY  kq  0.5 k!e , k P t0, 1, 2, . . . u
(a) (i)

PpY 2q  PpY ¤ 1q
 PpY  0q PpY  1q
0
0.5
0.51 e
0.5
 0.5 0!e 1!
 0.6065 0.3033
 0.9098
C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 50
Random Variable Tutorial Questions Set 8

(ii)

PpY ¡ 2q  1
PpY ¤ 2q
 1
rPpY  0q Pp Y  1q PpY  2qs
0.50 e
0.5 0.51 e
0.5 0.52 e
0.5
1
0! 1! 2!
 1
r0.6065 0.3033 0.0758s
 0.0144
(b) The time interval specified is one week. Since events in a Poisson distribution
are assumed to be independent,

Ppno accident in 3 weeksq  Ppno accident in one week  Ppno accident in one week 


Ppno accident in one week


 PpY  0q
3

 p0.6065q3
 0.2231

Tutorial Questions Set 8

Give answers to 4 decimal places.
1. If X ∽ P oisp3q, find:
(a) PpX  2q
(b) PpX  3q
(c) PpX ¥ 3q
(d) PpX ¤ 3q
2. If X ∽ P oispλq and PpX  4q  3P pX  3q, find λ and PpX  5q.
3. If X ∽ P oispλq and PpX  0q  0.323, find the value of λ to two
decimal places and use this to calculate PpX  3q

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 51

Random Variable 6.7 Hypergeometric Distribution

4. Is the following distribution a Poisson distribution?

x 0 1 2 3 4 5 6 ¥7
PpX  xq 0.3680 0.3679 0.1839 0.0613 0.0153 0.0031 0.0005 0.0000
5. Suppose that number of errors per page of a document follows a Poisson
distribution with mean λ and given the following table as the distribution of
errors per page of a document.
No of errors (x) 0 1 2 3 4 5 6 ¥7
No. of pages (f) 3 5 4 4 1 2 3 0
(a) Find the mean error per page
(b) Write the PMF of the distribution
(c) Find the probability that a page selected at random:
i. has no error
ii. has less than 2 errors
iii. has 3 or more errors
6. *Suppose that the probabilty p of an earthquate occuring in a year in a given
city is 0.025.
(a) Taking λ  np, find the probability that earthquake occurs exactly 2
times in n  150 years. Round to 3 decimal places.
(b) Show that the problem is an approximation of a binomial experiment
— where each year represents a trial, p is the probability of success and
n is the number of trials — by Poisson distribution. [Hint: Evaluate
the probability in part (a) with B p150, 0.025q and show that the result
is approximately equal to the result obtained by Poisson distribution in
part (a)]

6.7 Hypergeometric Distribution

The hypergeometric distribution is a discrete probability distribution.
Consider a finite population of N objects in which K objects are of the same kind.
Suppose we randomly draw n objects from the population without replacement.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 52

Random Variable 6.7 Hypergeometric Distribution

Let a success be the event that an item drawn belong to K. Then the distribu-
tion of the number of successes X is called the hypergeometric distribution with
parameters N , K, n.
Stated differently, the hypergeometric distribution describes the probability of ob-
taining k successes in n random draws from a population of size N wherein K
items are of the success state or feature.
A random variable X follows the hypergeometric distribution if its PMF is given
by
 
K N
K
n
k
PpX  kq  P t0, 1, 2, . . . , nu
k
 , k (6.12)
N
n

where,
N is the population size,
K is the number of success states in the population,
n is the number of draws/trials.
k
 is the number of observed successes,
n
is a binomial coefficient (defined here in subsection 6.5).
k
The hypergeometric distribution is the “drawing without replacement” counterpart
of the binomial distribution.
The mean and variance of the hypergeometric distribution with parameters N , K,
n are given, respectively, as
nK pN
K qpN
nq
EpX q  and VpX q 
nK
N N 2 pN
1q

Example 6.8 A bag contains 5 red balls and 4 blue balls. 3 balls are selected
randomly one after the other from the bag without replacement. What is the
probability of getting exactly 2 red balls?
Success here is picking a red ball. Hence,
K  5, n  3 N  9
Let X be the number of red balls selected in 3 trials. The PMF of X is given by
 
5 4
3
k
PpX  kq  P t0, 1, 2, 3u
k
 , k
9
3

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 53

Random Variable 6.7 Hypergeometric Distribution

 
5 4
3
2
6 PpX  2q  2

9
3
 
5 4

 29 1 Ñ p:q
3
 0.4762
Let us demonstrate how the use of permutation and combination from section 3
leads to this beautiful and easy-to-use PMF of hypergeometric distribution with
this example.
Now we first find all possible permutaions of the 3 balls drawn i.e. ordered
arrangements of the balls as follows: (let R = red and B = blue)
RRR RRB RBR BRR RBB BRB BBR and BBB
We can see that 2 red balls can be obtained in any of the following 3 ways
RRB or RBR or BRR
Recall that there is no replacement and so the events are dependent. Each
selection reduces the total number in the bag as well as the number of balls of
the type selected. For instance RRB means first pick is red (with prob = 5/9)
and second pick is red which was first picked (with prob = 4/8) and third pick is
blue not yet picked (with prob = 4/7). Continuing in this manner we evaluate
the conditional probabilities of the boxed expression as follows:

5 4 4

4 5 4

9 8 7
  
5 4 4
9 8 7
 
9 8 7

 4 9 5 8 4 7 4 9 5 8 4 7



54
987
4
 

54
 4 987 3

 p4  3q 9 5 8 4 7

5
 p4P 2q 9PP 2
3

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 54


2! 5C
 p2! 4C 2q 9
3! C 3
2
Since nP k  k! nC k . See summary point 3 of KEYNOTES 4
5
C 2 4C 2 2!2!
 9
C3
 3! ()

Now, the term in () that is different from the result (:) of our example is

C2 
4 2!2!
.
3!
Evaluating this term gives


4!  2!2!
 4!
 4P 1  4C 1


2!2! 3!

 3!

Substituting back 4C 1 for 4C 2  in () gives our desired result when

2!2!
3!
compared with p:q term of our Example 6.8. That is,
 
5 4
5 4
Ppgetting exacty 2 reds ballsq    0.4762
C2 C1 2 1
9

C3 9
3

Tutorial Questions Set 9

1. Let the random variable A denote the number of aces in a five-card hand
dealt from a standard 52-card deck.
(a) Give the probability mass function of A
(b) Find the probability that exactly 3 are aces.
(c) Find the probability that 2 or more are aces.
2. An aquarium contains 102 fish, 20 of which have been tagged by a researcher.
15 fish are randomly caught from the lake. Let Y be the number of tagged
fish in the researcher’s sample.
(a) Find the probability mass function of Y .
(b) Find the probability that exactly 5 are tagged.
(c) Find the probability that 3 or less are tagged.

55
BIBLIOGRAPHY BIBLIOGRAPHY

Bibliography
Blitzstein, J. K. and Hwang, J. (2019). Introduction to Probability (Second
Edition). Chapman & Hall/CRC, Boca Raton.
Kroese, D. and Chan, J. (2014). Statistical Modeling and Computation. Springer,
New York.
Kroese, D. P. (2018). A short introduction to probability.
https://people.smp.uq.edu.au/DirkKroese/asitp.pdf.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 56

Answers to Selected Tutorial Questions

Appendix

Figure 4: A deck of cards

Answers to Selected Tutorial Questions

Tutorial Questions Set 1 (p. 11)
tp4, 6q, p5, 5q, p6, 4q, p5, 6q, p6, 5q, p6, 6qu b.) (0,1000) or tx : 0 x 1000u
1. a. !
c. xi : 0 xi 200cm, i P t1, 2, 3, .., 10u d. tHHT, HT H, T HH, HHH u

2. a. t1H, 2H, 3H, 4H, 5H, 6H, 1T, 2T, 3T, 5T, 4T, 6T u b. tu or ϕ c. t6H, 6T u
d. (i) t1H, 1T, 2H, 2T, 3H, 3T, 5H, 5T u (ii) t3H, 3T, 5H, 5T u (iii) t2H, 2T, 4H, 4T, 6H, 6T u
(iv) t1H, 1T, 4H, 4T, 6H, 6T u (v) t1H, 1T, 2H, 2T, 4H, 4T, 6H, 6T u (vi) t4H, 4T, 6H, 6T u
e. (i) The event that an odd-numbered face or a prime-numbered face ap-
pears on the die. (ii) The event that a face both odd-numbered and prime-
numbered (i.e. odd-prime numbered face) appears on the die (iii) The event
that an odd-numbered face does not appear on the die. (iv) The event that
a prime-numbered face does not appear on the die. (v) The event that a
face both odd-numbered and prime-numbered does not appear on the die.
(vi) The event that an odd-numbered face or a prime-numbered face does
not appear on the die.

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 57

Answers to Selected Tutorial Questions

 c  c
¤
n £
n £
n ¤
n
4. Ai  Aci and  AciAi
i i i i
Alternatively, in expanded form: pA1 Y A2 Y    Y An qc  Ac1 X Ac2 X  X Acn
and pA1 X A2 X    X An qc  Ac1 Y Ac2 Y    Y Acn
Tutorial Questions Set 2 (pp. 21-23)
2 1 3 2 2 1 1 1 4
1. 0.26 2. 3. 4. 5. 6. 0.82 7. 8. 9. 10. 18% 11. 12.
7 2 5 3 7 49 256 221 663
16. PpAq 19. 0.7375 20. 0.8 21.
1 8 1 3
13. 14. 15.
114 16575 783 17

Tutorial Questions Set 3 (p. 29)

1. 6.84  10
6 , an almost impossible event 2. (a)

1 1 1 1
(b) (c) (d)
6 30 5 2
1
(e) 3. 0.1169 4. 10 5.
4 63 1
6. 7.
1
8.
45
2 1250 3 49 1024

Tutorial Questions Set 4 (p. 33)

1. 31, 24, 17, 7, 6, 12, 1, 38, 27, 25, 15, 18, 19, 34, 20, 48, 21, 36. 2. No, Yes, No,
Yes, Yes, No, No, No, Yes, Yes. Assumption: 1 represents a “Yes” and 0 represents
a “No” 4. 81, 77, 93, 94, 72, 96, 68, 82, 60, 98.

Tutorial Questions Set 5 (pp. 40-41)

1. (a) 0.9938 (b) 0.9878 (c) 0.3944 2. (a) 0.1587 (b) 0.8413 (c) 0.3413 3. (a) 0.5
(b) 0.9545 (c) 0.1587 4. (a) 0.9234 (b) 0.0766 (c) 0.0021

Tutorial Questions Set 6 (p. 44)

1. (a) 5 days (b) 25 days (c) (i) 0.6321 (ii) 0.0800 (iii) 0.0028 2. 0.0025 3. (a)
0.1054 (b) 0.5353 (c) 0.3127 4. 1.6, 2.56

Tutorial Questions Set 7 (pp. 48-29)

1 1
1. 0.3750 2. (a) 0.1762 (b) 0.0207 3. (a) (b)
8 4

Tutorial Questions Set 8 (pp. 51-52)

1. (a) 0.2240 (b) 0.2240 (c) 0.5768 (d) 0.6472 2. 12, 0.0127 3. 1.13, 0.0777

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 58

Answers to Selected Tutorial Questions

 x

57
exp
5722
(b) PpX  xq  , x P t0, 1, 2, . . . u (c) i. 0.0750
57 22
4. Yes 5. (a)
22 x!
ii. 0.2691 iii. 0.4793 6. (a) 0.1647

Tutorial Questions Set 9 (p. 55)

 
4 48
5
k
1. (a) PpA  k q  P t0, 1, 2, 3, 4u (b) 0.0017 (c) 0.0417
k
 ,k
52
5

C. O. Odijie STA110 (Introduction to Statistics) Lecture Notes Page 59

.

©2021 C. O. Odijie
All Rights Reserved.

These Lecture Notes are strictly for teaching purpose.

No part must be reproduced in any form
without the consent of the author.

Email: [email protected]