Chapter 1

Introduction
Lectures 1 - 3.
In this chapter we introduce the basic notions random experiment, sample space, events and probability
of event.
By a random experiment, we mean an experiment which has multiple outcomes and one don't know in
advance which outcome is going to occur. We call this an experiment with `random' outcome. We assume
that the set of all possible outcomes of the experiment is known.
Definition 1.1. Sample space of a random experiment is the set of all possible outcomes of the random
experiment.
Example 1.0.1 Toss a coin and note down the face. This is a random experiment, since there are
multiple outcomes and outcome is not known before the toss, in other words, outcome occur randomly.
More over, the sample space is

Example 1.0.2 Toss two coins and note down the number of heads obtained. Here sample space is
.

Example 1.0.3 Pick a point `at random' from the interval

picking any point. Sample space is

. `At random' means there is no bias in

Definition 1.2 ( Event ) Any subset of a sample space is said to be an event.

Example 1.0.4

is an event corresponding to the sample space in Example 1.0.1.

Definition 1.3 (mutually exclusive events) Two events

are said to be mutually exclusive if

.
If

and

are mutually exclusive, then occurrence of

Note that non occurrence of

Example 1.0.5 The events
But the events

implies non occurrence of

need not imply occurrence of

,

, since

and vice versa.

need not be

of the sample space in Example 1.0.1 are mutually exclusive.

are not mutually exclusive.

Now we introduce the concept of probability of events (in other words probability measure). Intuitively
probability quantifies the chance of the occurrence of an event. We say that an event has occurred, if the
outcome belongs to the event. In general it is not possible to assign probabilities to all events from the
sample space. For the experiment given in Example 1.0.3, it is not possible to assign probabilities to all
. So one need to restrict to a smaller class of subsets of the sample space. For the
subsets of
random experiment given in Example 1.0.3, it turns out that one can assign probability to each interval in
as its length. Therefore, one can assign probability to any finite union of intervals in
, by
representating the finite union of intervals as a finite disjoint union of intervals. In fact one can assign
probability to any countable union interval in
by preserving the desirable property "probability of
countable disjoint union is the sum of probabilities". Also note that if one can assign probability to an
event, then one can assign probability to its compliment, since occurence of the event is same as the
non-occurance of its compliemt. Thus one seek to define probability on those class of events which
satisfies "closed under complimentation" and "closed under countable union". This leads to the following
special family of events where one can assign probabilities.
Definition 1.4 A family of subsets
following.
(i)

of a nonempty set

is said to be a

-field if it satisfies the

(ii) if

, then

(iii) if

, then

Example 1.0.6

Let

Then

i.e.,

-fields. Moreover, if

is the smallest and

Then

Let

is a

Lemma 1.0.1

is a

is the largest

be a nonempty set and

-field and is the smallest

is called the

is a

be a nonempty set. Define

are

Example 1.0.7

-field generated by

Let

-field of subsets of

-field of subsets of

, then

. Define

-field containing the set

be an index set and

be a family of

-fields. Then

-field.

Proof. Since

for all

, we have

. Now,

Similarly it follows that

Hence

is a

-field.

Example 1.0.8

is a

Let

. Then

-field and is the smallest

-field containing

This can be seen as follows. From Lemma 1.0.1,

clearly

. If

-field containing

(ii) if

, then

-field containing

is a
, then

-field. From the definition of

. Hence,

Definition 1.5 A family

(i)

is a

. We denote it by

of subsets of a non empty set

is said to be a field if

is the smallest

(iii) if

, then

Example 1.0.9 Any

.
-field is a field. In particular,

Example 1.0.10 Let

Then

are fields.

. Define

is a field but not a

-field.

Note that (i) and (ii) in the definition of field follows easily. To see (iii), for
are finite so is
(iii) follows. i.e.,
To see that

and if either

or

, if both

is finite, then

is finite. Hence

is a field.

is not a

-field, take

Now

Definition 1.6 (Probability measure)

Let

be a nonempty set and

be a

-field of subsets of

. A map

is said to be

a probability measure if P satisfies

(i)
(ii) if

are pairwise disjoint, then

Definition 1.7 (Probability space).

; where

The triplet

, a nonempty set (sample space),

,a

-field and

, a probability

measure; is called a probability space.

. Define

on

as follows.

Example 1.0.11

Let

Then

is a probability space. This probability space corresponds to the random experiment of

tossing an unbiased coin and noting the face.

Example 1.0.12

Let

Then

is a probability space.

Solution.

. Define

on

as follows.

If

(This holds since

are pairwise disjoint. Then

's are disjoint)

Therefore

Therefore properties (i) , (ii) are satisfied. Hence

Theorem 1.0.1 (Properties of probability measure)
are in
(1)

. Then
.

(2) Finite sub-additivity:

(3)Monotonicity: if

, then

(4)Boole's inequality (Countable sub-additivity):

(5)Inclusion - exclusion formula:

(6)Continuity property:
(i) For

(ii) For

Proof. Since

is a probability measure.
Let

a probability space and

This proves (1). Now

Therefore

since

. This proves (3).

We prove (5) by induction. For

and

(Here

Hence we have

and

Combining the above, we have

Assume that equality holds for

Consider

Therefore the result true for

. Hence by induction property (5) follows.

From property (5), we have

Hence

Thus we have (2).

Now we prove (6)(i). Set

Then

are disjoint and

(1.0.1)
Also
(1.0.2)

Using (1.0.1), we get

Now using the definition of convergence of series, one has

(1.0.3)
Hence from (1.0.3), we have

Proof of (6)(ii) is as follows.

Note that

Now using (6)(i) we have

i.e.,

Hence

From property (2), it follows that

i.e.,

Therefore
(1.0.4)
Set

Then

and are in

. Also

Hence
(1.0.5)
Here the second equality follows from the continuity property 6(i). Using (1.0.5), letting
(1.0.4), we have

in

Recall that all the examples of probability spaces we had seen till now are with sample space finite or
countable and the

-field as the power set of the sample space. Now let us look at a random experiment

with uncountable sample space and the

-field as a proper subset of the power set.
Consider the random experiment in Example 1.0.3, i.e, pick a point 'at random' from the interval

Since point is picked 'at random', the probability measure should satisfy the following.
(1.0.6)
The

-field we are using to define P is

is called the Borel

-field of subsets of

Our aim is to define

, the

for all elements of

Clearly

-field generated by all intervals in

, preserving (1.0.6). Set

Let

. then

can be represented as

where

Then

where

Therefore

.
, it follows from the definition of

For
Hence
Define

that

is a field.
on

as follows.

(1.0.7)
where

's are pair wise disjoint intervals of the form

Extension of

from

to

follows from the extension theorem by Caratheodary. To understand the

statement of the extension theorem, we need the following definition.

Definition 1.8 (Probability measure on a field) Let

be a nonempty set and

be a field. Then

is said to be a probability measure on

if

(i)
(ii) if

Example 1.0.13

be such that

The set function

Theorem 1.0.2 (Extension Theorem)

extension to

are pairwise disjoint and

, then

given by (1.0.7) is a probability measure on the field

A probability measure defined on a field

has a unique

Using Theorem 1.0.2, one can extend

defined by (1.0.7) to

see Exercise 1.6 , there exists a unique probability measure

. Since

on

preserving (1.0.6).