Probability (Math 230)

Sample Space and sigma eld

Mujahid Abbas
Department of Mathematics, Room 129. [email protected]

First week, 2011

About this course:

Probability and Statistics for Engineers and Scientists, Walpole, Myers, Myers, Prentice-Hall.

About this course:

Probability and Statistics for Engineers and Scientists, Walpole, Myers, Myers, Prentice-Hall. Quizzes 20%

About this course:

Probability and Statistics for Engineers and Scientists, Walpole, Myers, Myers, Prentice-Hall. Quizzes 20% Midterm 35%

2 3

About this course:

Probability and Statistics for Engineers and Scientists, Walpole, Myers, Myers, Prentice-Hall. Quizzes 20% Midterm 35% Final Examination 45%

2 3 4

About this course:

Probability and Statistics for Engineers and Scientists, Walpole, Myers, Myers, Prentice-Hall. Quizzes 20% Midterm 35% Final Examination 45% There will be 12 quizzes and 10 best will be counted.

2 3 4 5

About this course:

Probability and Statistics for Engineers and Scientists, Walpole, Myers, Myers, Prentice-Hall. Quizzes 20% Midterm 35% Final Examination 45% There will be 12 quizzes and 10 best will be counted. There is no make-up for any quiz in any case

2 3 4 5 6

About this course:

Probability and Statistics for Engineers and Scientists, Walpole, Myers, Myers, Prentice-Hall. Quizzes 20% Midterm 35% Final Examination 45% There will be 12 quizzes and 10 best will be counted. There is no make-up for any quiz in any case All quizzes will be announced

2 3 4 5 6 7

About this course:

Probability and Statistics for Engineers and Scientists, Walpole, Myers, Myers, Prentice-Hall. Quizzes 20% Midterm 35% Final Examination 45% There will be 12 quizzes and 10 best will be counted. There is no make-up for any quiz in any case All quizzes will be announced All Examinations (Closed Book and Closed Notes)

2 3 4 5 6 7 8

Element of uncertainty:

It is likely to rain in the afternoon today

Element of uncertainty:

1 2

It is likely to rain in the afternoon today Perhaps I earn promotion next year

Element of uncertainty:

1 2 3

It is likely to rain in the afternoon today Perhaps I earn promotion next year My candidate stands a bright chance of winning the election.

Element of uncertainty:

1 2 3 4

It is likely to rain in the afternoon today Perhaps I earn promotion next year My candidate stands a bright chance of winning the election. The words "likely", "perhaps " "chance" convey the idea of uncertainty.

Element of uncertainty:

1 2 3 4

It is likely to rain in the afternoon today Perhaps I earn promotion next year My candidate stands a bright chance of winning the election. The words "likely", "perhaps " "chance" convey the idea of uncertainty. We cannot imagine what shape the human activities will have without the element of uncertainty.

Element of uncertainty:

1 2 3 4

It is likely to rain in the afternoon today Perhaps I earn promotion next year My candidate stands a bright chance of winning the election. The words "likely", "perhaps " "chance" convey the idea of uncertainty. We cannot imagine what shape the human activities will have without the element of uncertainty. In this course we shall try to measure this element of uncertainty.

What shall we learn today:

What is a sample space

What shall we learn today:

1 2

What is a sample space Examples

What shall we learn today:

1 2 3

What is a sample space Examples What are events

What shall we learn today:

1 2 3 4

What is a sample space Examples What are events Examples

What shall we learn today:

1 2 3 4 5

What is a sample space Examples What are events Examples Some suggested problems

What is an experiment

An experiment is a process that gives results

Denition A single performance of an experiment is called trial. The result of an experiment is called an outcome.

What is an experiment

1 2

An experiment is a process that gives results Any process/operation used to generate or obtain results/observations/data is referred to as an Experiment

Denition A single performance of an experiment is called trial. The result of an experiment is called an outcome.

Types of experiments:
There are two types of experiments 1 Deterministic experiment ( An experiment whose outcomes can be predicted with certainty). In these experiments conditions determine the outcomes. We associate the idea of certainty with such experiments.

Types of experiments:
There are two types of experiments 1 Deterministic experiment ( An experiment whose outcomes can be predicted with certainty). In these experiments conditions determine the outcomes. We associate the idea of certainty with such experiments. 2 We heat water to 100C under atmospheric pressure, water begins to boil. Under this condition no other outcome is possible.

Types of experiments:
There are two types of experiments 1 Deterministic experiment ( An experiment whose outcomes can be predicted with certainty). In these experiments conditions determine the outcomes. We associate the idea of certainty with such experiments. 2 We heat water to 100C under atmospheric pressure, water begins to boil. Under this condition no other outcome is possible. 3 A bag contains 8 black balls of same shape and size, one ball is drawn, it can not be green.

Types of experiments:
There are two types of experiments 1 Deterministic experiment ( An experiment whose outcomes can be predicted with certainty). In these experiments conditions determine the outcomes. We associate the idea of certainty with such experiments. 2 We heat water to 100C under atmospheric pressure, water begins to boil. Under this condition no other outcome is possible. 3 A bag contains 8 black balls of same shape and size, one ball is drawn, it can not be green. 4 Random experiment is an experiment whose outcomes can not be predicted. In well contested hockey match Pakistan is awarded a penalty corner. Will it be able to score a goal. In spite of thorough knowledge of the team capability for connecting the penalty corner and rival team capability of s defending it, we can not predict the outcome.

Sample Space

When one conducts a random experiment, uncertainty jumps in. First, we shall try to list or know the entire set of possibilities. The rst ingredient of our study is a set dened as follows: Denition The set consisting of all possible outcomes of a random experiment is called a sample space or possibility space. It is denoted by S. Thus sample space gives information concerning the outcomes of an experiment. We can associate more than one samples spaces with the same random experiment. The generic symbol for an outcome of a random experiment is .

Sample Space

Denition Each element of a sample space is called a sample point. Denition If a sample space is a set of nite number of elements or an unending sequence with as many elements as there are natural numbers then it is called a discrete sample space otherwise it is a continuous sample space.

Examples:

Example Tossing of a coin is a random experiment. For one coin toss the set S = fH, T g represents the sample space, where T stands for a tail and H stands for a head. Example Suppose we have an initial stock price S, at next period stock price may become dS (downward movement) or uS (upward movement), (0 < d < 1 < u).

Example

Example Suppose two electric bulbs are tested. A bulb may be defective (D) or non defective (N ). The sample space will be the following set S = f(N, N ), (D, D ), (N, D ), (D, N )g

Example

Example Suppose that investor has to invest in two stocks: A and B. Each stock may increase in value (I ), decrease in value (D ) and remains unchanged in value (U ). Consider an experiment as the investment in two stocks. Sample space in this case will be S

= f(I , I ), (I , D ), (I , U ), (D, I ), (D, D ), (D, U ), (U, I ), (U, D ), (U, U )g

Some more examples:

Example Consider an experiment of rolling a pair of die, the set of all possible outcome is 2 3 (1, 1) (2, 1) (3, 1) (4, 1) (5, 1) (6, 1) 6 (1, 2) (2, 2) (3, 2) (4, 2) (5, 2) (6, 2) 7 6 7 6 (1, 3) (2, 3) (3, 3) (4, 3) (5, 3) (6, 3) 7 6 7 6 (1, 4) (2, 4) (3, 4) (4, 4) (5, 4) (6, 4) 7 6 7 4 (1, 5) (2, 5) (3, 5) (4, 5) (5, 5) (6, 5) 5 (1, 6) (2, 6) (3, 6) (4, 6) (5, 6) (6, 6) In some experiments it is helpful to list the elements of the sample space systematically by means of tree diagram.

More Examples:

Suppose we toss a coin repeatedly and count number of tosses required until rst head appears. In this case the sample space will be S = f1, 2, 3, ...g.

Sample space with large or innite number of elements is best described by a statement or a rule.

More Examples:

If we ask a person to pick a point from the interval [0, 1], then this experiment has uncountable many outcomes and the sample space in this case will be S = [0, 1]

Suppose we toss a coin repeatedly and count number of tosses required until rst head appears. In this case the sample space will be S = f1, 2, 3, ...g.

Sample space with large or innite number of elements is best described by a statement or a rule.

More Examples:

If we ask a person to pick a point from the interval [0, 1], then this experiment has uncountable many outcomes and the sample space in this case will be S = [0, 1] The sample space of experiment of measuring the life time of transistor will be S = [0, )

Suppose we toss a coin repeatedly and count number of tosses required until rst head appears. In this case the sample space will be S = f1, 2, 3, ...g.

Sample space with large or innite number of elements is best described by a statement or a rule.

More Examples:

If we ask a person to pick a point from the interval [0, 1], then this experiment has uncountable many outcomes and the sample space in this case will be S = [0, 1] The sample space of experiment of measuring the life time of transistor will be S = [0, ) If we throw a dart at a circular target of radius 3 inches, assuming our dart always land on the dart board. Our sample space in this case will be S = f(x, y ) : x 2 + y 2 9g.

Suppose we toss a coin repeatedly and count number of tosses required until rst head appears. In this case the sample space will be S = f1, 2, 3, ...g.

Sample space with large or innite number of elements is best described by a statement or a rule.

Event
Denition If S is the sample space of an experiment, then an event is any subset of the sample space. An event that contains exactly one sample point is called a simple event or elementary event. Denition An event that contains more than one sample points is called a compound event. It is produced by the union of simple events. Since a sample space is a subset of itself, therefore it is also an event called certain event. Denition An empty set is also a subset of S, it is called an impossible event. Denition If the outcome of an experiment is contained in an event, then we

Example:
Example If we roll a die to see which face lands on top then this experiment has six possible outcomes and the resulting sample space is S = f1, 2, 3, 4, 5, 6g if we are interested to see if number on the top face is even then S1 = f2, 4, 6g if we are interested to see if number on the top face is odd then S2 = f1, 3, 5g Note that S1 and S2 are contained in S. Hence S provides more information than S1 and S2 .

Venn diagram

The relationship between events and the corresponding sample space can be illustrated graphically by means of Venn diagrams. Sample space is represented by rectangle and events are represented by the circles inside the rectangle.

Example:

Example Consider the experiment of tossing a coin repeatedly. Let A be the event that number of tosses required until rst head appears is less than 5. Then A = f1, 2, 3, 4g We shall say that an event A has occurred if an outcome is an element of set A.

Algebra of Events:

We can use the usual set operations to obtain new events: If A and B are events associated with some random experiment then
1

A [ B stands for the event A or B occurs and it consists of all outcomes that are either in A or in B or in both A and B.

Algebra of Events:

We can use the usual set operations to obtain new events: If A and B are events associated with some random experiment then
1

A [ B stands for the event A or B occurs and it consists of all outcomes that are either in A or in B or in both A and B. A \ B stands for the event A and B occurs. It consists of all outcomes that are both in A and B.

Algebra of Events:

We can use the usual set operations to obtain new events: If A and B are events associated with some random experiment then
1

A [ B stands for the event A or B occurs and it consists of all outcomes that are either in A or in B or in both A and B. A \ B stands for the event A and B occurs. It consists of all outcomes that are both in A and B. A0 or Ac , the complement of A, is the set of all outcomes in the sample space which are not in A (stands for the event A does not occur)

Denitions:

Two events that have no outcome in common are said to be mutually exclusive or disjoint. In other words, A \ B = AB = = fg. (They can not both occur simultaneously )

Denitions:

Two events that have no outcome in common are said to be mutually exclusive or disjoint. In other words, A \ B = AB = = fg. (They can not both occur simultaneously ) if A and B are events associated with some random experiment the occurrence of an event A implies the occurrence of an event B, it is represented by A B

Some Rules:

if A and B are events of some random experiment then we have the following rules:
1

(A [ B )c = Ac \ B c and (A \ B )c = Ac [ B c (Demorgan s law)

Some Rules:

if A and B are events of some random experiment then we have the following rules:
1

(A [ B )c = Ac \ B c and (A \ B )c = Ac [ B c (Demorgan s law) A \ Ac = and A [ Ac = S

Some Rules:

if A and B are events of some random experiment then we have the following rules:
1

2 3

(A [ B )c = Ac \ B c and (A \ B )c = Ac [ B c (Demorgan s law) A \ Ac = and A [ Ac = S S c = and c = S

Examples:

Example A box contains 3 marbles, 1 red, 1 green and 1 blue. Consider an experiment that consists of taking 1 marble from the box and then replacing it in the box and drawing a second marble from the box. Describe the sample space. Repeat when the second marble is drawn without rst replacing the rst marble. S

= f(r , r ), (r , g ), (r , b ), (g , g ), (g , r ), (g , b ) (b, b ), (b, g ), (b, r )g

in second case S = f(r , g ), (r , b ), (g , r ), (g , b ), (b, g ), (b, r )g

Example:
A hospital administrator codes incoming the patient suering gunshot wounds according to whether they have insurance (coding 1, if they do and 0 if they do not) and according to their condition which is rated as good (g), fair (f) and serious (s). Consider an experiment that consists of the coding of such a patient then S

= sample space of this experiment = f(1, g ), (0, g ), (1, f ), (0, f ), (1, s ), (0, s )g

A = event that patient is in serious condition = f(0, s ), (1, s )g B = event that patient is uninsured = f((0, g ), (0, f ), (0, s )g B c [ A = f(1, g ), (1, f ), (1, s ), (0, s )g

Sigma eld:

We may be interested in several events with respect to the same experiment. We collect all the events into a class. Let S be any nonempty set ( sample space in our case ) and F a collection of subsets of S. F is called sigma eld if it satises the following conditions:
1

and S belong to F

Sigma eld:

We may be interested in several events with respect to the same experiment. We collect all the events into a class. Let S be any nonempty set ( sample space in our case ) and F a collection of subsets of S. F is called sigma eld if it satises the following conditions:
1 2

and S belong to F If A is in F then Ac is in F

Sigma eld:

We may be interested in several events with respect to the same experiment. We collect all the events into a class. Let S be any nonempty set ( sample space in our case ) and F a collection of subsets of S. F is called sigma eld if it satises the following conditions:
1 2 3

and S belong to F If A is in F then Ac is in F If A1 , A2 , A3 , ... are in F then [i 1 Ai is in F =

Continued:

Typically, we take F to be all subsets of S ( power set of S ). Elements of F are called events. Elements of S are denoted by . it is a consequence of the second property that the empty set is an element of every sigma eld. It is then a consequence of the third property and DeMorgan Laws that the intersection of a countable s number of elements of the sigma eld is in the sigma eld. An ordered pair (S, F ) is called a measurable space.

Examples:

In case of throwing a coin, sample space is S = fH, T g and F1 = f, S,fH g, fT gg and F2 = f, S g are sigma elds.

Examples:

In case of throwing a die, sample space is S = f1, 2, 3, 4, 5, 6g. F1 = P (S ) = the collection of all subsets of S ( 26 = 64 elements set) , F2 = f, S,f1, 3, 5g, f2, 4, 6gg, F3 = f, S g are sigma elds are sigma elds on S. Note that sigma elds represent information ( in sigma eld F2 , f1, 2, 3g is not an event).

In case of throwing a coin, sample space is S = fH, T g and F1 = f, S,fH g, fT gg and F2 = f, S g are sigma elds.

Examples:

The smallest sigma eld which contains all open intervals in R is called Borel sigma eld and is denoted by B(R) . Its elements are called Borel sets.

In case of throwing a die, sample space is S = f1, 2, 3, 4, 5, 6g. F1 = P (S ) = the collection of all subsets of S ( 26 = 64 elements set) , F2 = f, S,f1, 3, 5g, f2, 4, 6gg, F3 = f, S g are sigma elds are sigma elds on S. Note that sigma elds represent information ( in sigma eld F2 , f1, 2, 3g is not an event).

In case of throwing a coin, sample space is S = fH, T g and F1 = f, S,fH g, fT gg and F2 = f, S g are sigma elds.

Examples:

If C is any class of subsets of S, there is a smallest sigma eld in S containing C ( called sigma eld generated by C )

Examples:

If C is any class of subsets of S, there is a smallest sigma eld in S containing C ( called sigma eld generated by C ) It is also intersection of all sigma elds containing C .

Some suggested problems:

Questions 04, 07, 08, 12, and 19 ( pages 29-31)