65 views

Uploaded by Anna Xi

- Roussas g a Course in Mathematical Statistics
- MIT Lecture Notes on Probability
- MIT6_436JF08_lec05
- Worksheet
- Urgente - Calculadora Estadistica en Visual c Sharp
- Statistics Tutorial
- r059210401 Probability Theory and Stochastic Process
- MCA4020_Summer 2016 Assignment
- Modeling Uncertainty With Probability
- Probability and Stochastic Processes
- Discrete Random Variables.ppt
- Discrete math
- Random Variables.docx
- 0104215v1
- SkriptNMS
- Random Variables Without Joint Distributions
- MTE-03
- Fall07 Notenotes for pee seogrts
- Curso-Basico-JKSimMet
- PPT Independent and Dependent

You are on page 1of 32

TO

PROBABILITY

1|Page

Introduction to Probability

2|Page

3. INTRODUCTION TO PROBABILITY

OBJECTIVES

On completion of this unit students should be able to:

1. Calculate and interpret simple probabilities using counting and tree diagrams

2. Calculate probabilities of mutually exclusive events, independent events and complementary

events

3. From a crosstabulation calculate conditional probabilities

4. Use the probability distribution of a discrete random variable to calculate probabilities and

to calculate and interpret the expected value and standard deviation of the random variable.

3.1

Prelude

Roger Federer is to play Andy Roddick. Who will win? If you select Federer it is unlikely that you will

be totally certain of his victory but rather you would say that he would probably win, or that he had a

better (or higher) chance than Roddick of winning.

The topic of probability aims to quantify this sense of probably or more likely or higher chance. If we

say that 100% probability means that a player is certain to win and 0% means that they are certain to

lose then 50% would mean that both players have the same chance. If you select Federer to win, what

percentage probability would you allocate?

60%

80%

95%

quite likely to win, only a small chance of losing

almost certain to win

There are many other situations where the outcome is uncertain but where a probability can be assigned.

Many of these are subjective so that different people will give different estimates of the probability.

Examples include sporting contests, whether it will rain tomorrow and share prices rising or falling.

In other situations the probability can be calculated exactly by simple counting methods and appropriate

assumptions. For example in tossing a fair coin the probability of a head is 50% or 0.5. This is because

there are only two equally likely outcomes. We would say that the chance is 1 out of 2, or 1/2 = 0.5.

This example suggests a simple definition of probability in the case where all events are equally likely:

Pr(event) =

total number of possible events

Example 3-1

In one roll of a fair 6 sided die (singular of dice) what is:

(a) Pr(6)

(b)

Pr(even number)

(c) Pr(number less than 5)

(d)

Pr(2 and 4)?

Solution

(a)

(b)

(c)

(d)

Pr(6) = 1/6

Pr(even number) = 3/6 =0.5

Pr(number < 5) = 4/6 = 2/3

Pr(2 and 4) = 0/6 = 0

as there are 3 even number faces

as there are four numbers less than 5

as 2 and 4 cannot occur together.

3|Page

Introduction to Probability

Note that, as in the above example, probabilities are positive proper fractions, that is probabilities are

numbers between 0 and 1 (inclusive). However in everyday language we may also convert them to

percentages.

3.2

Tree Diagrams

As events become more complicated we need some special methods to help us count the number of

different events. A simple visual method is the tree diagram. For tossing a fair coin the diagram looks

like this:

The total number of events equals the number of branches at the end of the (horizontal!) tree. For a fair

coin this is 2.

Example 3-2

Draw a tree diagram for the random event of rolling a fair die.

Solution

Tree diagrams are especially useful when two or more events take place together or successively.

Example 3-3

Two fair coins are tossed. How many outcomes are there? How many of these have 1 head and 1 tail?

Solution

4|Page

There are four outcomes, namely: {H,H}, {H,T}, {T,H} and {T,T}. Of these, two have one head and

one tail. Hence

Pr(1 head and 1 tail) = 2/4 = 0.5

Note that in this example the sum of the four probabilities, each equal to one quarter (= 0.25), is 1. This

is always true for the complete collection of outcomes that make up one random experiment.

Basic properties of probabilities:

(i) probabilities are greater than or equal to zero and less than or equal to 1, ie 0 pi 1

(ii) the sum of all possible probabilities is one, ie

pi = 1 .

3.3

Another source of exact probabilities is frequency tables.

Example 3-4

One hundred people are surveyed to identify their bank. From the following table what is the probability

that a person selected at random from the 100 people is a customer of the Comm bank?

Bank

Bonza

Comm

Natty

Melb

Total

No. of

customers

20

26

34

20

100

Solution

Pr(Comm) = 26/100 = 0.26, as each person has the same chance of selection in random sampling.

Hence each person is equally likely to be selected.

3.4

There are several ways in which two events can be related. The ideas will be illustrated using the simple

random experiments of coin and dice and also tabulations of data. These relationships apply to any type

of probabilities including subjective and exact probabilities.

3.4.1

The complement of an event is the collection of all the other possible events. So in one toss of a fair

coin the complement of a head is a tail; for one roll of a fair die the complement of the even numbers {2,

4, 6} is the set of odd numbers {1, 3, 5}. Since the sum of the probabilities of all events equals 1 we

have the connection that

Pr(event) + Pr(complement) = 1.

5|Page

Introduction to Probability

Example 3-5

For the bank customer data in Example 3-4 find the complement of:

(a) {Bonza} and (b) {Bonza, Melb} and the probabilities of these complements.

Solution

(a)

The

complement

of

{Bonza}

then Pr({Comm, Natty, Melb})

(b)

{Comm,

Natty,

=

=

=

=

1

1 Pr({Bonza})

1 20/100

0.8

Melb}.

Pr({Comm, Natty})

3.4.2

is

=

=

=

1 Pr({Bonza, Melb})

1 40/100

0.6

Two events are mutually exclusive if they have no events in common. This is equivalent to saying that

if one event occurs it is impossible for the other event to occur (at the same time). For example in

rolling a fair die these pairs of events are mutually exclusive:

{even numbers} and {odd numbers}.

However the following two events are not mutually exclusive: {1, 2, 3, 4} and {2, 4, 6}, as they have

both 2 and 4 in common. This means that if, with one roll of the die, we get, say, a 4 then both events

have occurred.

In Australian Rules football scoring has two mutually exclusive outcomes: a goal and a behind. (A

behind is a near miss of the goal on either side.)

Example 3-6

Which of the following are mutually exclusive pairs of events?

(a)

(b)

(c)

(d)

(e)

(f)

Living in Geelong and working in Melbourne.

Being under 21 years of age and being Prime Minister of Australia.

Being a parliamentarian and being a farmer.

Drawing a red card or drawing an Ace out of an ordinary deck of playing cards.

Forming a five person committee that contains 2 women or that contains three

women.

Solution

(a), (c) and (f) are mutually exclusive pairs of events.

If two events are mutually exclusive then Pr(either one or the other occurs) = Pr(one) + Pr(other). If we

denote two events using the letters A and B then if they are mutually exclusive

Pr(A or B) = Pr(A) + Pr(B)

6|Page

Example 3-7

For one roll of a fair die, what is the probability of a 2 or a 4?

Solution

We can calculate Pr(2 or 4) as 2/6 by the counting method, as there only two numbers in the event. But

since {2} and {4} are mutually exclusive and Pr(2) = 1/6 = Pr(4) then

Pr(2 or 4) = 1/6 + 1/6 = 2/6

In the bank data of Example 3-4, as each of the 100 customers belongs to only one bank the total

frequency is 100 and belonging to a bank gives four mutually exclusive events. Example 3-4(b) we can

calculate the probabilities in this way:

and

Pr({Comm or Natty}) = 26/100 + 34/100 = 0.6

Consider the following table. One hundred people were asked which charities they supported.

Charity

Frequency

Salvation Army

44

Smith Family

35

Heart Foundation

22

Deaf and Blind

24

Guide Dogs

28

total

153

As the total is more than 100 some people are supporting more than one charity. Hence this is not a set

of mutually exclusive events and it would be incorrect to say, for example,

Pr({Salvation Army or Smith Family})

= 44/153 + 35/153

as some people are counted twice in this calculation.

3.4.3

Independent Events

This is a quite different type of relationship to mutually exclusive as the property of independence is

conferred by the physical method of performing the events. The basic method for guaranteeing

independence is to select the items at random. This means that the selection of any one particular item

has no influence on what other items will be selected.

Some simple examples of random events that are independent are:

Tossing a fair coin several times. Each toss is (physically) independent of previous tosses as

these cannot affect the outcome of any later toss. So if the first toss is a head on the second toss

heads and tails are equally likely.

Rolling two fair dice together. Whatever comes up on one face cannot affect the other. Of

course the two dice may collide but this can only lead to an accidental change in the numbers.

There is no systematic effect such as if one is an even number then the other must be an odd

number.

7|Page

Introduction to Probability

Two weeks Tattslotto numbers. The numbered balls are thoroughly mixed and each one is

selected at random. The numbers selected in one week have no effect on which numbers are

selected in subsequent weeks.

Using a computer program that generates random numbers, to select individuals for a survey.

This method is used to randomly select telephone numbers for opinion polls.

Example 3-8

Which of the following pairs of events are independent and which are not independent?

(a)

(b)

(c)

(d)

Having a flat tyre and being late for work.

Living in Hawthorn and being over 2 metres tall.

A student studying hard and passing an exam.

Solution

(a) and (c) are independent. (b) and (d) are (probably) dependent.

Example 3-9

In one toss of a fair coin we get 10 straight heads. On the next toss is tails more likely than heads?

Solution

No! As each toss is independent of all other tosses then heads and tails are equally likely at every toss,

regardless of what has happened before.

If two events are independent then the probability that both happen together can be calculated from their

individual probabilities. Call two events A and B. If A and B are independent events then:

Pr(A and B) = Pr(A).Pr(B)

Example 3-10

Two fair coins are tossed together. What is the probability that both coins show a head?

Solution

Since Pr(head) = 0.5 for each coin, and the coins are tossed independently of each other then:

Pr(2 heads) = 0.5x0.5 = 0.25.

This agrees with, and supplements, the tree diagram analysis in example 3.3.

3.4.4

Conditional Probability

If two events are not independent then they are dependent. In this case the probability of one event

occurring is affected by whether or not the other related event has occurred. Related events are often

displayed in crosstabulations where two variables are measured for a set of individuals.

8|Page

Example 3-11

A Bondi Beach nightclub has the following data on the age and marital status of 140 customers:

Age

Single

Under 30

77

30 and over

28

14

21

Marital Status

Married

(a)

(b)

(c)

(d)

(e)

Comment on the age of customers attending the club.

Comment on the marital status of customers attending the club.

What is the probability of finding a customer who is single and under the age of 30?

If a customer is under 30, what is the probability that he or she is single?

Solution

In this table we have recorded the age and marital status of 140 people. Each cell gives a different

combination of age and marital status. For example there are 14 people aged under 30 who are married.

(a)

Age

Single

Under 30

0.55

30 and over

0.20

0.10

0.15

Marital Status

Married

(b)

(c)

(d)

(e)

(f)

This question asks for a conditional probability. The condition is that the customer is under

30. This means that we must go back to the original table and only consider the under 30

column of the age variable.

Age

Single

Under 30

77

Marital Status

Married

14

There are 91 customers under 30. Hence the (conditional) probability that a customer is single, given

that they are under 30, is 77/91 = 0.846. Since the unconditional probability that a customer is single is

0.75 (from (c)) this shows that being single and being under 30 are not independent as these two

probabilities are not equal.

9|Page

Introduction to Probability

3.5

Example 3-12

A companys employees have been classified according to age and salary as shown below:

< $25,000

< 30

30 to 45

>45

TOTAL

32

10

1

43

$25,000 to

$45,000

3

18

10

31

> $45,000

TOTAL

0

21

5

26

35

49

16

100

One employee is selected at random and two events are defined as follows:

A: The employee is under 30

B: The employees salary is under $25,000.

Express each of the following probabilities in words and find their numerical values:

(a) Pr(A)

(b) Pr(B

(c) Pr(A and B)

(d) Pr(A or B)

(e) Pr(A occurs given that B occurs)

(g) Find the probability that the employees salary is at least $25,000 given that the employee is at least

30 years of age.

(h) Are events A and B mutually exclusive?

(i) Are events A and B independent events?

Solution

(a) Pr(A) = 35/100 = 0.35

$25,000) is 0.32

= 46/100 = 0.46

$25,000) is 0.46

= 0.744

also under 30) is 0.744

= 0.914

$25,000) is 0.914

= (18 + 21 + 10 + 5)/(10 + 18 + 21 + 1 + 10 + 5)

= 54/65

= 0.831

(h) A and B are not mutually exclusive as an employee can be both.

(i) From (a) and (b), Pr(A)xPr(B) = 0.35x0.43 = 0.151, which is different from (c). Hence A

and B are not independent.

10 | P a g e

Example 3-13

Note: The solutions to each part of this question will be given immediately after the question part.

Bendigo Power Generation Pty Ltd (BPG) is currently starting work on a project designed to increase

the generating capacity of one of its plants in the Bendigo District of Central Victoria. The project is

divided into two sequential stages: stage 1 (design) and stage 2 (construction). While each stage will be

scheduled and controlled as closely as possible, management cannot predict beforehand the exact

elapsed time for each stage of the project. An analysis of similar construction projects over the past 3

years has shown completion times for the design stage of 2, 3, or 4 months and completion times for the

construction stage of 6, 7, or 8 months.

(a) Draw a tree diagram for the BPG project, including a listing of all possible outcomes.

(b) Based on their experience and judgment, management concluded that the experimental outcomes

were not equally likely. By conducting a study of the completion time for similar projects

undertaken by BPG over the past three years management found the data on the next page.

Using this data, calculate the probabilities associated with each sample point and add these

probabilities to the tree diagram (on next page).

Completion time (months)

Stage 1

Stage 2

Sample Point

2

2

2

3

3

3

4

4

4

6

7

8

6

7

8

6

7

8

(2,6)

(2,7)

(2,8)

(3,6)

(3,7)

(3,8)

(4,6)

(4,7)

(4,8)

completion times

6

6

2

4

8

2

2

4

6

TOTAL = 40 projects

11 | P a g e

Introduction to Probability

Stage 1

Design

Stage 2

Construction

6

7

8

2

0.15

0.15

0.05

3

0.10

0.20

0.05

4

0.05

0.10

0.15

(c) Due to the critical need for additional power, management has set a goal of 10 months for the total

project completion time. Hence the entire project will be completed late if the total elapsed time to

complete both stages exceeds 10 months. What is the probability that the project will be finished on

time, that is, in 10 months or less?

Pr(on time) = (0.15 + 0.15 + 0.05) + (0.10 + 0.20) + 0.05 = 0.70

We can add these probabilities because they are all mutually exclusive outcomes.

(d) What is the most likely or most probable completion time?

Pr(8 months) = 0.15

Pr(11 months) = 0.15

Pr(12 months) = 0.15

(e) What is the probability that the project will be completed within a year?

Pr(completed within a year) = 1 because the largest possible prediction time is 12 months.

3.6

3.6.1

Random Variables

In the previous sections and exercises the examples were concerned with random events and the

probabilities of the different outcomes of these events. For example in one toss of a fair coin there were

two outcomes, Head and Tail, each with probability 0.5; in Example 3-4 people were customers of one

of four banks and the probability of being a customer of each bank was calculated from the frequencies.

12 | P a g e

In all these examples we have two things: a random variable, namely the outcome of a random

experiment, and a set of probabilities corresponding to each of the outcomes. When we combine these

two we have the probability distribution of the random variable.

In our examples in this section probability distributions will be displayed as a table, as in Example 3-14.

This is because all the random variables are discrete ie the outcomes are all distinct numbers or

categories. (Random variables can also be continuous, in which case a graph or table is used to

represent the probability distribution. An important example is discussed in section 4 below.)

Random variables are given a single letter name such as X or Y. Thus in one toss of a fair coin X is the

outcome, either a head or a tail.

Example 3-14

(a) What is the probability distribution for one toss of a fair coin?

Tossing a fair coin

x

Pr(x)

Head

0.5

(b) What is the probability distribution for one roll of a fair die?

Rolling a fair Die

1

2

x

1/6

1/6

Pr(x)

Tail

0.5

3

1/6

4

1/6

5

1/6

6

1/6

What is the probability of getting a number greater than 4? Name the probability rule that you need

to calculate this probability.

Pr(X > 4) = Pr(5) + Pr(6) = 1/6 + 1/6 = 2/6 = 1/3. Since 5 and 6 are mutually exclusive events we

can add the individual probabilities.

(c) Students in a university residence visit a nearby Pizza Parlour with varying frequencies per month.

Let X be the number of times a student visits the Pizza Parlour each month. The probability

distribution of X could look like this:

x

Pr(x)

0

0.1

1

0.3

2

0.4

3

0.2

If a student is selected at random find the probability that he/she will visit the Pizza Parlour:

i) No times.

Pr(X = 0) = 0.1

ii) Once.

Pr(X = 1) = 0.3

are mutually exclusive.

Pr(X) = 1, i.e. the sum of all the probabilities is one, and

0 Pr(X) 1 for all values of X.

13 | P a g e

Introduction to Probability

3.6.2

We can make a graph of a probability distribution by putting the X values on the horizontal axis and the

probabilities on the vertical axis.

Example 3-15

Graphs of two of the probability distributions from Example 3-14, (a) and (c).

(a)

(c)

In these graphs it is the top of the vertical lines that represents the probabilities. The rest of the line is

drawn only to make the point more visible.

3.6.3

Given a set of data we can calculate the sample mean and standard deviation. Similarly, given a

probability distribution we can calculate the mean and standard deviation of the corresponding random

variable. The formulae and procedures are very similar to those for sample data and can be carried out

for discrete distributions using Microsoft Excel.

The mean of a numerical random variable X is also called the Expected Value of X, written E(X). The

name derives from the idea that the mean is the value of X that would be expected as the most likely or

average value in the long run. It is calculated from the formula:

= E ( X ) = X . Pr( X )

The variance of a numerical random variable X is often written Var(X). For a discrete random variable,

Var ( X ) = E ( X 2 ) 2

The standard deviation is

Var ( X )

14 | P a g e

Example 3-16

Calculate the mean value of the result of one roll of a fair die.

Solution

1

6

1

6

1

6

1

6

1

6

= E ( X ) = X . Pr( X ) = 1 + 2 + 3 + 4 + 5 + 6

1 21

=

= 3.5

6 6

The average of all the faces is 3.5, which can be seen by adding the numbers from 1 to 6 and then

dividing by 6.

Example 3-17

What is the average number of times per month that a student (from Example 3-14(c)) goes to the Pizza

Parlour?

Solution

The average number of times per student is the expected value of X.

ie a student selected at random is expected to go to the Pizza Parlour on average 1.7 times per

month.

The expected value and variance of a discrete random variable can be calculated using Microsoft Excel

as in the following example.

Example 3-18

Calculate the mean and standard deviation of the Pizza Parlour data of Example 3-14(c).

Solution

expected value, variance and standard deviation of

a discrete random variable

Step 1:

Enter

the

probability

distribution

Enter the values of X and Pr(X) given on p.13 so that your spreadsheet

appears as shown below.

15 | P a g e

Introduction to Probability

Step 2:

Calculating the

expected value

To calculate the expected value we need to first multiply the outcome (X)

of each event in column 1 by its probability Pr(X) in column 2 of the

spreadsheet. Therefore, go to cell C2 and type = a2*b2 and press ENTER.

To copy this formula to the other cells in column C, right-click cell C2 and

select Copy from the resulting drop-down menu. Then press the Shift key,

scroll down with the left mouse button so that C3 to C5 are also

highlighted, right-click the mouse again, and select Paste.

Your

spreadsheet should then appear as shown below.

We now need to add the values in column C since the expected value is the

sum

X . Pr( X )

We will do this in row 7 of the spreadsheet here. First, type E(X) in cell

A7 so we can keep track of our calculations. Then, in cell B7, type

=sum(c2:c5) and press ENTER, to obtain the result below.

16 | P a g e

Step 3:

Calculating

E( X 2 )

to

To calculate X^2 for the first outcome (X) in row 2, go to cell D2 and type

=a2^2 and press ENTER.

Then copy this formula to the other cells in column D by right-clicking cell

D2 and selecting Copy from the resulting drop-down menu. Press the Shift

key, scroll down with the left mouse button so that D3 to D5 are also

highlighted, right-click the mouse again, and select Paste.

column D by the corresponding cell in column B. Therefore, go to cell E2

and type =d2*b2 and press ENTER. Then copy and paste this formula to

cells E3 through E5 so that your spreadsheet appears as below.

in cell B8, we will sum the X 2 * P( X ) values. That is, we will sum cells

E2 to E5. Hence, in cell B8, type = sum(e2:e5) and press ENTER.

Hence E ( X 2 ) = 3.7

17 | P a g e

Introduction to Probability

Step 4:

calculating the

variance

We will now calculate Var(X) in row 9, so type Var(X) in cell A9. Since

Var ( X ) = E ( X 2 ) 2 and we calculated E ( X 2 ) in cell B8 and in cell

B7, go to cell B9 and type =b8-b7^2 and press ENTER.

Step 5:

Calculating the

standard

deviation

First, type Standard Deviation in cell A10. The standard deviation is the

square root of the variance (which we calculated in cell B9), which can be

calculated using the sqrt function in Excel. That is, go to cell B10 and type

=sqrt(b9) and press ENTER.

Example 3-19

A company operates a chain of takeaway food outlets. It intends to build an additional outlet and

surveys the following possible locations:

Ringwood

Footscray

St. Kilda

Probability of

Success

0.6

0.4

0.5

Estimated Profit

if Successful

$35,000

$60,000

$45,000

Estimated Profit

if Unsuccessful

$2,500

$7,000

$6,000

Which location should be chosen? Consider both the expected profit and the risk involved as measured

by the standard deviation.

18 | P a g e

Solution

For each location we can calculate the expected value of the profit from the probability distribution. For

Ringwood the distribution is:

x (profit $)

35,000

2,500

0.6

0.4

Pr(x)

The probability of failure is 1 0.6 = 0.4 as failure is the complement of success.

For Footscray the distribution is:

x (profit $)

Pr(x)

60,000

0.4

7,000

0.6

45,000

0.5

6,000

0.5

x (profit $)

Pr(x)

We obtain the three expected profits and standard deviations individually, using Microsoft Excel as in

Example 3-18.

Location

Ringwood

Footscray

St.Kilda

22,000

28,200

25,500

15,922

25,965

19,500

Coeff of variation

72%

92%

76%

Hence we could choose Footscray as it is the most profitable location but noting that the expected profit

is the most variable and hence the most risky. St Kilda is a compromise with the second highest

expected profit but the second lowest risk as measured by the coefficient of variation.

19 | P a g e

Introduction to Probability

3.7

References:

Anderson, D.R, et al Statistics for Business and Economics, Australian edition 1989.

Chapter 4, pages 79 - 107

Chapter 5, pages 125 - 132

Levine, D. M. Krehbiel, T. C., Berenson, M. L./ Business Statistics: A First Course, Second Edition

2000, Prentice Hall.

Chapter 4, pages 154 170, 179 - 183

McLean, A., Stephens, B., Business Mathematics and Statistics, Longman 1996.

Chapter 20, pages 451 477

Chapter 21, pages 488 494

Selvanathan, A., Selvanathan, S., Keller, G., Warrack, B., Australian Business Statistics, Abridged third

edition, Nelson 2004.

Chapter 4, pages 114 140

Chapter 5, pages 182 190

Chapter 19, pages 453 - 475.

Probability

Tree diagram

Basic requirements of probability

Mutually exclusive

Independent events

Conditional probability

Random variable

Probability distribution

Expected value

Event

Definition of probability of an event

Complement

Addition Law

Multiplication Law

Discrete random variable

Probability function

Standard deviation

USEFUL FORMULAE:

Pr(event ) = 1 Pr(complement )

E(X) = = xp(x)

Pr(A and B) = Pr(A)Pr(B), for independent events

20 | P a g e

3.8

Exercises

Exercise Set 3A Calculating Probabilities (Solutions can be found on page

266)

1. Over a long period of time, the queue length of customers at the teller section of a major bank was

observed to have the following probability distribution;

Number in queue

Probability

0

0.1

1

0.2

2

0.2

3

0.3

4 or more

0.2

(a)

At most two people in the queue.

(b)

(c)

(d)

(e)

Given there is at least one person in the queue, what is the probability that there are 4 or

more?

2. An airline has offered a special vacation package to Fiji. The length of stay is either for 3 days or 7

days, and the type of accommodation can be either economy, regular, or deluxe. Consider the

experiment of observing the choices made by the next person signing up for the package.

(a)

(b)

3. An investor has two stocks: stock A and stock B. Each stock may increase in value, decrease in value,

or remain unchanged. Consider the experiment as the investment in both of the two stocks.

(a)

(b)

(c)

How many of the experimental outcomes result in an increase in value for at least one of the

two stocks?

(d)

How many of the experimental outcomes result in an increase in value for both of the

stocks?

4. Suppose that a manager of a large apartment complex provides the following subjective probability

estimates about the number of vacancies that will exist next month:

Vacancies

0

1

2

3

4

5

Probability

.05

.15

.35

.25

.10

.10

21 | P a g e

Introduction to Probability

List the components of each of the following events and calculate the probability of the events:

(a)

No vacancies.

(b)

(c)

5. The manager of a furniture store sells from zero to four china cabinets each week. Based on past

experience, the following probabilities are assigned to sales of zero, one, two, three, or four cabinets:

Pr(0) = 0.08

Pr(1) = 0.18

Pr(2) = 0.32

Pr(3) = 0.30

Pr(4) = 0.12

1.00

(a)

(b)

Let A be the event that two or fewer are sold in one week. Find Pr(A).

(c)

Let B be the event that four or more are sold in one week. Find Pr(B).

6. A sample of 100 customers of the Adelaide Gas and Electric Company resulted in the following

frequency distribution of monthly charges.

Amount $

0-49

50-99

100-149

150-199

200-249

Number

13

22

34

26

5

(a)

Let A be the event that monthly charges are $150 or more. Find Pr(A).

(b)

Let B be the event that monthly charges are less than $150. Find (Pr(B).

7. A survey of 50 students at a technical college regarding the number of extracurricular activities resulted

in the following data

Number of Activities

Frequency

0

8

1

20

2

12

3

6

4

3

5

1

(a)

Let A be the event that a student participates in at least 1 activity. Find Pr(A).

(b)

Let B be the event that a student participates in 3 or more activities. Find Pr(B).

(c)

22 | P a g e

8. A garage sells from 0 to 4 car batteries each week. Based on past experience, the following probabilities

are assigned to sales of 0, 1, 2, 3 or 4 batteries.

(a)

Pr(0) = 0.19

Pr(1) = 0.24

Pr(2) = 0.35

Pr(3) = 0.14

Pr(4) = 0.08

1.00

Are these valid probability assignments? Why or why not?

(b)

Let A be the event that two or fewer are sold in one week. Find Pr(A).

(c)

Let B be the event that four or more are sold in one week. Find Pr(B).

(d)

Are A and B mutually exclusive? Find Pr(A and B) and Pr(A or B).

9. A purchasing agent has placed two rush orders for a particular raw material from two different suppliers,

A and B. If neither order has arrived in 4 days the production process must be shut down until at least

one of the orders arrives. The probability that supplier A can deliver the material in 4 days is 0.55. The

probability that supplier B can deliver the material in 4 days is 0.35. Draw a tree diagram to help you

answer the following questions.

(a)

What is the probability that both suppliers deliver the material within 4 days? Since two

separate suppliers are involved we are willing to assume independence.

(b)

What is the probability that the production process is shut down in 4 days because of a

shortage in raw material (that is, both orders are late)?

(c)

What is the probability that at least one supplier delivers the material in 4 days?

10. To install a pool fence, a company responds to all sales inquiries by first sending a brochure. It then

attends the property to measure and quote, and finally installs the fence. It classifies the outcome of all

sales inquiries into three mutually exclusive categories depending on how far the customer progresses in

the inquiry:- request for brochure; measure and quote given; and pool fence purchased. From its records

for the last 6 months it has extracted the result of all inquiries and the suburb of the customer. The

results are summarised in the table below.

Request

for

brochure

Measure

and

quote

Fence purchased

Total

Eastern suburbs

100

Western suburbs

50

Other

50

Total

200

50

10

20

80

50

200

0

60

20

90

70

350

(a)

What is the probability that a sales contact results in a pool fence sale?

(b)

What is the probability that a sales inquiry from a person in the Eastern suburbs results in a

pool fence sale?

(c)

What is the probability that a measure and quote inquiry comes from the Eastern or Western

suburbs?

(d)

Are the events 'Request for brochure' and 'client is from the Eastern suburbs' independent?

Give a reason for your answer.

23 | P a g e

Introduction to Probability

Exercise Set 3B Discrete Probability Distributions (Solutions can be found on page 30)

1. The heights of the Hawthorn Basketball Club team playing next week (including injury replacements)

are:

Height (cm)

No. of Players

170

1

175

2

180

2

185

2

190

1

195

4

200

1

205

2

210

1

(a)

(b)

(c)

Find the probability distribution of this random variable. Determine the expected height of a

Hawthorn Basketball Club team member using Microsoft Excel.

2. The demand for a product of Gippsland Industries varies greatly from month to month. Based on the

past 2 years of data, the following probability distribution shows the companys monthly demand.

300

0.20

Unit demand

Probability

400

0.30

500

0.35

600

0.15

(a)

If the company places monthly orders based on the expected value of the monthly demand,

what should Gippslands monthly order quantity be for this product?

(b)

Assume that each unit demanded generates $70 in revenue and that each unit ordered costs

$50. How much will the company gain or lose in a month if it places an order based on your

answer to part (a) and where the actual demand for the item is 300 units?

(c)

3. An investor has a certain amount of money available to invest now. Three alternative portfolio

selections are available. The estimated profits of each portfolio under three economic conditions are

shown in the table below.

Event

Economy declines

No change

Economy expands

A

$500

$1,000

$2,000

Portfolio Selection

B

-$2,000

$2,000

$5,000

(a)

(b)

(c)

C

-$7,000

-$1,000

$20,000

Probability

0.3

0.5

0.2

24 | P a g e

4. Consider the state of the economy. There are three possible changes in the economy between this year

and next. For each type of change the probability and likely return are given in the following table.

State of Economy

Recession

Stable

Boom

Probability of State

p(x)

0.25

0.5

0.25

(x)

-0.05

0.15

0.35

5. A state lottery is conducted in which 10,000 tickets are sold for $1 each. Six winning tickets are

randomly selected: one grand prize winner of $5,000, one second prize winner of $2,000, one third prize

winner of $1,000 and three other winners of $500 each.

(a)

Construct a probability distribution, with the random variable (X) equivalent to the net

return.

(b)

(c)

6. The demand for a dried fish product which is a specialty of Tran Trung Industries varies greatly from

month to month. Based on the past three years of data, the following probability distribution shows the

companys monthly demand.

Demand (kgs)

Probability

200

0.15

300

0.30

400

0.40

500

0.15

(a)

If the company prepares monthly production on the basis of the expected value of the

monthly demand, what quantity of dried fish should Tran Trung Industries prepare?

(b)

Assume that each kilogram of fish demanded generates $25 in revenue and that each

kilogram of dried fish produced costs $17. How much will the company gain or lose if its

monthly production is based on your answer from (a) and the actual demand for the product

is 300 kgs?

7. The J.R. Ryland Computer Company is considering a plant expansion that will enable the company to

begin production of a new computer product. The companys managing director must determine

whether to make the expansion a medium or large-scale project. An uncertainty involves the demand for

the new product, which for planning purposes may be low demand, medium demand, or high demand.

The probability estimates for the demands are .20, .50 and .30, respectively. Letting X and Y indicate

the annual profit in $1,000s, the firms planners have developed profit forecasts for the medium and

large-scale expansion projects.

Expansion Profits

Demand

Low

Medium

High

Medium-scale

x

Pr(x)

50

0.20

150

0.50

200

0.30

Large-scale

y

Pr(y)

0

0.20

100

0.50

300

0.30

25 | P a g e

Introduction to Probability

(a)

Compute the expected value for the profit associated with the two expansion alternatives.

Which decision is preferred for the objective of maximising the expected profit?

(b)

Compute the standard deviation for the profit associated with the two expansion alternatives.

Which decision is preferred for the objective of minimising the risk or uncertainty?

ANSWERS TO EXERCISES

Exercises 3A Answers

1. (i)

(ii)

Pr(X 3) = Pr(X 2) + Pr(3) = 0.5 + 0.3 = 0.8 OR Pr(X 3) = 1 Pr( 4) = 1 0.2 = 0.8

(iii)

OR Pr(X 1) = Pr(1) + Pr(2) + Pr(3) + Pr( 4) = 0.2 + 0.2 + 0.3 + 0.2 = 0.9

(iv)

(v)

Pr(X 4 | X 1) =

Pr( X 4) 0.2

= 0.22

=

Pr( X 1) 0.9

2. (a)

Number of outcomes = No. of nights stay No. of accommodation types = 2 3 = 6

possible outcomes

(b)

26 | P a g e

3. (a)

(b)

(c)

(d)

1 = (,)

4. (a)

Pr(X = 0) = 0.05

(b)

(c)

5. (a)

Yes. All the probabilities are between 0 and 1 (0 pr(x) 1)and the sum of the

probabilities is 1 (pr(x) = 1).

(b)

(c)

6. (a)

(b)

26 + 5 31

=

= 0.31

100

100

27 | P a g e

Introduction to Probability

7. (a)

Pr(A) = Pr(X 1) =

20 + 12 + 6 + 3 + 1 42

=

= 0.84

50

50

(b)

Pr(B) = Pr(X 3) =

6 + 3 + 1 10

=

= 0.2

50

50

(c)

Pr(X = 2) =

12

= 0.24

50

8. (a)

Yes. All the probabilities are between 0 and 1 (0 pr(x) 1)and the sum of the

probabilities is 1 (pr(x) = 1).

(b)

(c)

(d)

Pr(A and B) = 0

Pr(A or B) = Pr(A) + Pr(B) = 0.78 + 0.08 = 0.86

9.

(a)

(b)

(c)

Pr(At least one on time) = Pr(OT, OT) + Pr(OT, L) + Pr(L, OT) = 0.1925 + 0.3575 + 0.1575

= 0.7075

OR Pr(At least one on time ) = 1 Pr(both late) = 1 0.2925 = 0.7075

28 | P a g e

10.

(a)

(b)

(c)

(d)

Let Pr(A) = Pr(Request for a brochure) = 200/350 = 0.571

Let Pr(B) = Pr(Client from Eastern suburbs) = 200/350 = 0.571

Thus, Pr(A) Pr(B) = 0.571 0.571 = 0.326

Now, Pr(A and B) = 100/350 = 0.286

Since Pr(A and B) Pr(A) Pr(B)Pr(A) i.e. 0.286 0.326 the events are not

independent.

29 | P a g e

Introduction to Probability

EXERCISE 3B Answers

Probability distribution.

Height

170

175

180

185

190

195

200

205

210

(cm)

Probability 0.0625 0.125 0.125 0.125 0.0625 0.25 0.0625 0.125 0.0625

Enter the height and the corresponding probabilities in two columns. To calculate the expected

value we need to first multiply the height of each in column 1 by its probability in column 2 of

the spreadsheet. Therefore, go to cell C2 and type = a2*b2 and press ENTER

Thus the expected height of a Hawthorn basketball club member is 190 centimetres.

a

b

Expected value=445

Revenue = 300 @ $70

= $21,000

Cost = 445 @ $50 = $22,250

Profit/Loss

= ($1,250)

Standard deviation=97.34

30 | P a g e

Highest expected return are Portfolios A and B, with an expected return of $1,400.

Lowest risk (as measured by the standard deviation) is Portfolio A with $522.

It depends on what one is looking for in an investment. Portfolio B has a good return and

medium risk, while Portfolio A has minimum risk but a lower return. Portfolio C offers a

good return with very high risk.

Thus the expected return is 15% and the standard deviation is 14% (Note the coefficient of

variation = 0.1414/0.15 = 94.27%, thus there is a great deal of uncertainty associated with

the returns).

Net return

$ (x)

Pr(x)

4999

1999

999

499

-1

0.0001

0.0001

0.0001

0.0003

0.9994

31 | P a g e

Introduction to Probability

Expected loss is 5 cents. A person is unlikely to win, but if one does, there are good returns.

A conservative gambler may be comfortable buying a ticket for such a small outlay (but

would be more cautious if the ticket cost say $100).

Expected value=$355

b

= $7,500

Cost = 355 @ $17 = $6,035

Profit/Loss

= $1,465

Higher expected return is for the medium scale expansion, with an expected return of

$145,000.

The standard deviations are $52,200 and $111,360 respectively. Lowest risk (as measured

by the standard deviation) is the medium scale expansion. Most of the times we should

calculate Coefficient of Variation but in this case expected returns for each investment are

close hence we are looking at the standard deviations.

32 | P a g e

- Roussas g a Course in Mathematical StatisticsUploaded byKarlilla Cuéllar
- MIT Lecture Notes on ProbabilityUploaded byOmegaUser
- MIT6_436JF08_lec05Uploaded byMarjo Kaci
- WorksheetUploaded byJohn Paul Ore
- Urgente - Calculadora Estadistica en Visual c SharpUploaded byIgor S Monrroy Peralta
- Statistics TutorialUploaded byMehraj Ahmed
- r059210401 Probability Theory and Stochastic ProcessUploaded byandhracolleges
- MCA4020_Summer 2016 AssignmentUploaded byAppTest PI
- Modeling Uncertainty With ProbabilityUploaded byMohammad Raihanul Hasan
- Probability and Stochastic ProcessesUploaded byAbdiasis Abdallah Jama
- Discrete Random Variables.pptUploaded byAnonymous QcSAY0p
- Discrete mathUploaded byKhairah A Karim
- Random Variables.docxUploaded bymeowskiee
- 0104215v1Uploaded byVigneshRamakrishnan
- SkriptNMSUploaded byasadullahj_1
- Random Variables Without Joint DistributionsUploaded bySalvador Priego Sánchez
- MTE-03Uploaded byVipin
- Fall07 Notenotes for pee seogrtsUploaded byAnoop Sree
- Curso-Basico-JKSimMetUploaded byXia ZYong
- PPT Independent and DependentUploaded byerikacarino
- Data Prep - Concentartion of Values (Excel)Uploaded bySpider Financial
- Lecture 1Uploaded byCarlos Andrés
- Lecture01 02 Intro Probability TheoryUploaded bySara Afzal
- MATH230 Lecture NotesUploaded byFoo
- Supplemental Material 2017Uploaded byTina Chen
- Eco assignmentUploaded byAnonymous ydR7QouJe
- uji tUploaded bySani Kartika Elf'suju
- Lecture 14Uploaded byVaibhav Shukla
- Statistics PresentationUploaded byNida Amri
- ENGG953 Spring2016 Lectures Lecture Bayesian Theorem Posterior ProbabilityUploaded byNasir Naeem

- Subordinate Demonstrative Science in the Sixth Book of Aristotle's PhysicsUploaded bydobizo
- V F DUploaded bychethan
- UMTS Indoor PlanningUploaded bytinhdv1982
- New Holland Brand Identity GuidelinesUploaded bysaumyavish
- Hospital Management SoftwareUploaded bysaurabhumalkar
- rsyslogUploaded byPichai Ng-arnpairojh
- 6. Financial + Salient FeatureUploaded bySUDHANSHU
- MB0049 Solved SMU Assignment MBAUploaded byArvind K
- Training Man-E Cisco_Multicast.pdfUploaded byTruong Mai Quang Anh
- Star StudiesUploaded byfedericovitella
- Development of Detection Maturity Method in Carrot Varieties Cipanas Using _Wiliandi Saputro and Irsyad Faisal MutaqiUploaded byWiliandi Saputro
- Issue-IUploaded byAnonymous 5uPEDI6
- Physical.opticsUploaded byKumar Raja
- accesorios bticinoUploaded byjczaurus
- My Script Elc501 ForumUploaded byAiedah Oto
- ATC DigitalUploaded bypmads07
- AIALATAPInnovationGlennHickman030410Uploaded bysteve7857
- Davao City Overland Transport Terminal - Organizational Set-upUploaded byMelodyFrancisco
- Lecture 6. Geo ReferencingUploaded byRazi Baig
- Cf RemoverUploaded byNielz Tulalessy
- Douglas 2000Uploaded byMade Deddy
- Determine Effect of Trench Excavation on Nearby Buildings by Rule of ThumbUploaded byLava Sat
- The INDEPENDENT Issue 532Uploaded byThe Independent Magazine
- LAB 04 - Gauss Jordan and Inverse of Matrices in MATLABUploaded byCan Raps
- Bro_14Uploaded bySailesh Bond
- QA Serbia - TA Fundamentals (3) - CopyUploaded byNenad Petrövic
- Bab 2.4 Elektronik Design and Technology Living SkillsUploaded byCik Tyqah
- plotting_workflow.pdfUploaded bymarian07k
- VP Operations Job DescriptionUploaded byAmol Ghemud
- beckerUploaded byfrenki06