STAT7055: LECTURE 2

Probability

Admin Stuff

Everyone should be enrolled in lectures and tutorials. If you have any lecture or tutorial enrolment issues, e.g., timetable clashes, please see Patricia Penm in the school office (4th floor CBE building) and bring your timetable with you.

STAT7055 Financial Statistics Lecture 2

Admin Stuff

Dont forget to regularly check Wattle for announcements, lecture notes, tutorial questions and solutions, etc. Please also regularly check your ANU email addresses. All announcements made on Wattle will also be sent to your ANU email addresses.
STAT7055 Financial Statistics Lecture 2

Terminology

A random experiment is a process that results in a number of possible outcomes, none of which can be predicted with certainty. Examples include:

Rolling a die: Possible outcomes are 1, 2, 3, 4, 5 or 6. Tossing a coin: Possible outcomes are heads or tails.
STAT7055 Financial Statistics Lecture 2 4

Terminology

The sample space of a random experiment is a list of all possible outcomes. Usually denoted by . For example, when rolling a die, the sample space is: = {1,2,3,4,5,6}. Outcomes must be mutually exclusive and exhaustive.

Mutually exclusive: No two outcomes can both occur at the same time on any one trial. Exhaustive: All possible outcomes must be included in the sample space.
STAT7055 Financial Statistics Lecture 2 5

Terminology

A simple event is an individual outcome in a sample space. Usually

denoted by or .

theif sample have space possible denoted by outcome So we = {1,2,,}.

STAT7055 Financial Statistics Lecture 2

Terminology

occurring on a single trial is written

Two important rules:

2.

STAT7055 Financial Statistics Lecture 2

Three Approaches to Assigning Probabilities

The classical approach is based on the assumption that the outcomes of an experiment are equally likely to happen. The classical probability utilises rules and laws. The relative frequency approach defines probability as the long-run relative frequency with which an outcomes occurs. The subjective approach is based on personal judgment, accumulation of knowledge, and experience.
STAT7055 Financial Statistics Lecture 2 8

Classical Approach

Suppose we roll a die once. Q: What is the probability we get a 5? Possible outcomes are = {1,2,3,4,5,6}. Assume all six outcomes are equally likely.

A: Probability we get a 5 is 1/6.

STAT7055 Financial Statistics Lecture 2 9

Relative Frequency Approach

I decide to start a door-to-door recruitment for ANU undergraduate students. We sample 200 final year undecided high school graduates, and 120 say yes to my sales pitch. Q: As I approach the house of an undecided HS grad, what is the probability that I am successful? A: Probability I am successful is 120/200=0.6.
STAT7055 Financial Statistics Lecture 2 10

Subjective Approach

Suppose I am an investor and Im trying to decide whether or not to invest in a particular stock. Q: What is the probability that this stock will increase in value? A: Based on past experience, my knowledge of the stock market and this particular stock, I assign a probability of 0.3.
STAT7055 Financial Statistics Lecture 2 11

Probability of an Event

An event is a collection of one or more simple (individual) events.

For example, when rolling a die, let denote the event that an odd number comes up. Then = {1,3,5}.

The probability of a favourable event occurring is equal to: number of favourable outcomes total number of outcomes
STAT7055 Financial Statistics Lecture 2

So () = 3/6 = 0.5.

12

Terminology and Notation

We can also combine events. There are some important combinations of events that we will encounter repeatedly throughout the course. Suppose we have two events, and .

For example, when rolling a die, let = {1,3,5} denote getting an odd number and = {3} denote getting a 3.
STAT7055 Financial Statistics Lecture 2 13

Terminology and Notation

Intersection:

; and ; intersection with . It means and both occur. E.g., = {3}.

Union:

; or ; union with . It means occurs, occurs or both occur. E.g., = {1,3,5}.

STAT7055 Financial Statistics Lecture 2

14

Terminology and Notation

Complement:

; complement. It means does not occur or the opposite of . E.g., = {2,4,6}.

Conditional:

|; given . It means occurs given that has occurred.

STAT7055 Financial Statistics Lecture 2

15

Important Facts Regarding Probability

For Something has to happen. = 1. the sample space , ()

For All probabilities must lie between 0 and 1. any event , 0 1.

If and are mutually exclusive, then = 0.

= , the empty set. E.g., is 2. They cant both1 and is the event we roll a the event we roll a happen.
STAT7055 Financial Statistics Lecture 2

Joint Probabilities

Suppose we are investigating the relationship between how well a mutual fund performs and where the fund managers earns his or her MBA. Let events be known as follows:

1 = Fund manager graduated from a top20 MBA program. 2 = Fund manager did not graduate from a top20 MBA program. 1 = Fund outperforms the market. 2 = Fund does not outperform the market
STAT7055 Financial Statistics Lecture 2

17

Joint Probabilities Continued

Probabilities :MutualFund 1 outperforms Market P(and)=0.11 P(and)=0.06 :MutualFund doesnotoutperform 2 Market P(and)=0.29 P(and)=0.54

:Top-20MBA :NotTop-20MBA

Joint probabilities: P(Mutual fund P(Mutual fund P(Mutual fund P(Mutual fund

STAT7055 Financial Statistics Lecture 2

18

Marginal Probabilities
Probabilities :MutualFund outperforms1Market :MutualFund doesnot 2 outperformMarket Totals

:Top-20MBA
:NotTop-20MBA Totals

P(and)=0.11
P(and)=0.06 P()=0.17 1

P(and)=0.29
P(and)=0.54 P()=0.83 2

P()=0.40
P()=0.60 1.00

Marginal probabilities:

Computed by adding across rows or down columns. Named because they are calculated in the margins of the table. P(1) = P(1 and 1) + P(1 and 2) = 0.11 + 0.29 = 0.40 P(2) = P(2 and 1) + P(2 and 2) = 0.06 + 0.54 = 0.60

P(1) = P(1 and 1) + P(1 and 2) = 0.11 + 0.06 = 0.17 P(2) = P(2 and 1) + P(2 and 2) = 0.29 + 0.54 = 0.83
STAT7055 Financial Statistics Lecture 2

19

Conditional Probability

Conditional probability: the probability of one event given the occurrence of another event. Definition (Bayes Theorem):
=

STAT7055 Financial Statistics Lecture 2

20

Conditional Probability Example

Probabilities :MutualFund outperforms1Market :MutualFund doesnot 2 outperformMarket Totals

:Top-20MBA
:NotTop-20MBA Totals

P(and)=0.11
P(and)=0.06 P()=0.17 1

P(and)=0.29
P(and)=0.54 P()=0.83 2

P()=0.40
P()=0.60 1.00

What is the probability that a mutual fund with a manager who graduated from a top-20 MBA program outperforms market? 1 1 = 1 and 1 1 0.11 = = 0.275 0.4
21

STAT7055 Financial Statistics Lecture 2

Conditional Probability Example

Probabilities :MutualFund outperforms1Market :MutualFund doesnot 2 outperformMarket Totals

:Top-20MBA
:NotTop-20MBA Totals

P(and)=0.11
P(and)=0.06 P()=0.17 1

P(and)=0.29
P(and)=0.54 P()=0.83 2

P()=0.40
P()=0.60 1.00

For a randomly selected mutual fund which underperforms the market, What is the probability that a graduate of a top-20 MBA program manage it? 1 2 = 1 and 2 2 0.29 = = 0.3494 0.83
22

STAT7055 Financial Statistics Lecture 2

Independence

Two events are independent if:

(|) = () or (|) = ()

In other words, two events are independent if the probability of one event is not affected by the occurrence of the other event.

If and are independent, then:

and = () = and / = / = ()

The above ONLY holds if and are independent.

STAT7055 Financial Statistics Lecture 2 23

Example

A store sells 2 brands of a particular product one expensive and the other inexpensive. Let expensive brand be called A, inexpensive brand be called B. A survey of 1000 sales gives the following: Brand

GenderA Male132

BTotal 147279

Female516
Total648

205721
3521000
STAT7055 Financial Statistics Lecture 2

24

Example
GenderA
Male132 Female516 Total648 Brand BTotal 147279 205721 3521000

P(Customer is male) = 279/1000 = 0.279 P(Purchase brand A) = 648/1000 =0.648 P(Purchase brand A AND female) = 516/1000 = 0.516

STAT7055 Financial Statistics Lecture 2

25

Example
Brand

GenderA
Male132 Female516 Total648

BTotal
147279 205721 3521000

P(Purchase brand A GIVEN female) = P(Purchase brand A AND female)/P(female) = (516/1000)/(721/1000) = 516/721
STAT7055 Financial Statistics Lecture 2

26

Example Are Gender and Brand Purchased Independent?

Brand

GenderA
Male132 Female516 Total648

BTotal
147279 205721 3521000

If independent: P(Purchase brand A GIVEN female) = P(Purchase brand A) LHS = 516/721 RHS = 648/1000 LHS RHS Therefore gender and brand purchased are NOT independent!
STAT7055 Financial Statistics Lecture 2 27

Multiplication Rule

and = = () Why? From Bayes Theorem for conditional probability we know:

() ()

and

() = ()

If and are independent, the multiplication rule directly gives us:

( and ) = ()()
STAT7055 Financial Statistics Lecture 2

Example Selecting Two Students Without Replacement

A finance course has 7 male and 3 female students. The lecturer wants to select two students at random to be his research assistants. What is the probability that the two students chosen are female?

Solution: Let represent the event that the first students event is the and chosen is thechosen W representalso female. that female second find student joint probability: ( and ) = (|)()
STAT7055 Financial Statistics Lecture 2 29

Example Selecting Two Students Without Replacement

The probability that the first student chosen is female is () = 3/10. What is the probability that the second student chosen is also female? Given that the first female student has been chosen, there are only 2 female students left and there are only 9 students left in the class. Therefore, (|) = 2/9. Thus, the joint probability is:

and = = 2 9 10
STAT7055 Financial Statistics Lecture 2

Example Selecting Two Students With Replacement

A finance course has 7 male and 3 female students. The lecturer will be away for two classes (2 weeks). Professor Jane will be their replacement for the next two classes. Jane will select one student at random in each class and pick on him or her to answer the questions. What is the probability that the two students chosen are female?

STAT7055 Financial Statistics Lecture 2

31

Example Selecting Two Students With Replacement

Solution: Let represent the event that the first students chosen is female and is also female.event that to second student chosen represent the We want the find the joint probability: = ()

However, it is now possible to choose the same student in the each of the two classes. Thus and are independent events, so we apply the multiplication rule for independent events. ( and ) = ()() = (3/10)(3/10) = 0.09
STAT7055 Financial Statistics Lecture 2

Addition Rule

Why? Let A be pink and B be blue.

If A and B are mutually exclusive events then

= +
STAT7055 Financial Statistics Lecture 2

Example Applying the Addition Rule

In Canberra, 22% of the households subscribe to the Canberra Times and 35% subscribe to the Australian. A survey reveals that 10% of all households subscribe to both newspapers. What is the probability that a randomly selected household subscribe to at least one of the newspapers? Let:

A = subscribes to the Canberra Times. B = subscribes to the Australia.

Solution: or = + and = 0.22 + 0.35 0.1 = 0.47

STAT7055 Financial Statistics Lecture 2 34

Complement Rule

STAT7055 Financial Statistics Lecture 2

35

Complement Rule

Since ( = 1, we know that: = = 1 Using the Addition Rule for mutually exclusive events we also know that:

= +

Therefore + = 1. Interpretation: Either happens or it doesnt.

STAT7055 Financial Statistics Lecture 2

Example Coin Tossing

An unbiased coin is tossed twice.

= {,,,}; all equally likely (prob=1/4) Find the probability of at least 1 tail. Let event = At least 1 tail = {,,}.

() = 3/4.
Alternatively: = 1 = 1 no tails = 1
1 = 4 3 4

STAT7055 Financial Statistics Lecture 2

37

Worked Examples

Q6.13 (pg. 180). A survey asks adults to report

their marital status. The sample space is = {single, married, divorced, widowed}. Use set notation to represent the event that the adult is not married.

Let = {married}. Therefore: = {unmarried} = {single, divorced, widowed}

STAT7055 Financial Statistics Lecture 2

38

Worked Examples

Q6.14 (pg. 180). Suppose that in the city in which the survey is conducted, 50% of adults are married, 15% are single, 25% are divorced, and 10% are widowed. Assignment probabilities to each simple event in the sample space. Which approach did you use in Part (a)?

a)

b)

STAT7055 Financial Statistics Lecture 2

39

Worked Examples

Solution: P(married) = 0.5 P(single) = 0.15 P(divorced) = 0.25 P(widowed) = 0.1 Relative frequency approach. Why?

a)

b)

STAT7055 Financial Statistics Lecture 2

40

Worked Examples

a) b) c)

Q6.15 (pg. 180). Find the probability of each of the following events. The adult is single. The adult is not divorced. The adult is either widowed or divorced.
Solution: P(single) = 0.15. P(divorced) = 0.25. Therefore P(not divorced) = 1-0.25 = 0.75.

a) b)

STAT7055 Financial Statistics Lecture 2

41

Worked Examples
c)

P(widowed or divorced) = P(widowed) + P(divorced) P(widowed & divorced) Since widowed and divorced are mutually exclusive, then P(widowed & divorced) = 0.

Therefore: P(widowed or divorced) = 0.1 + 0.25 = 0.35

STAT7055 Financial Statistics Lecture 2

42

Worked Examples

Q6.59 (pg. 198). A study undertaken by the Miami-dade Supervisor of Elections in 2002 revealed that 44% of registered voters are Democrats, 37% are Republicans, and 19% are Others. If 2 registered voters are selected at random, what is the probability that both of them have the same party affiliation?

Solution:
Let = {Same Party} = {,,}. Then: = 0.44 (0.44) = 0.1936 = 0.37 (0.37) = 0.1369 = 0.19 (0.19) = 0.0361
STAT7055 Financial Statistics Lecture 2 43

Worked Examples

Therefore, the probability that both of the registered voters have the same party affiliation is: = + + = 0.1936 + 0.1369 + 0.0361 = 0.3666

STAT7055 Financial Statistics Lecture 2

44

Worked Examples

Q6.77 (pg. 208). Bad gums may mean a bad heart. Researchers discovered that 85% of people who have suffered a heart attack had periodontal disease. Only 29% of healthy people have this disease. Suppose that in a certain community heart attacks occur with 10% probability. If someone had periodontal disease, what is the probability that he or she will have a heart attack?

STAT7055 Financial Statistics Lecture 2

45

Worked Examples

Lets define some events:

Let = have periodontal disease.

Let = have suffered a heart attack.

We know from the question that = 0.1, = 1 = 0.9, = 0.85 and = 0.29. We want to find (|).
STAT7055 Financial Statistics Lecture 2

46

Worked Examples

Using the definition of conditional probability: =

We need to find ( ) and (). For we can use the Multiplication Rule: = = 0.85 0.1 = 0.085
STAT7055 Financial Statistics Lecture 2

47

Worked Examples

For (), we know from marginal probabilities that: = + ( )

For we again use the Multiplication Rule: = = 0.29 0.9 = 0.261

Therefore, = 0.085 + 0.261 = 0.346. Finally, = / = 0.085/0.345 = 0.246.

STAT7055 Financial Statistics Lecture 2

Readings

Chapter 6.

Keller, 9th Edition.

STAT7055 Financial Statistics Lecture 2

49