1.

6 POSTERIOR PROBABILITIES 55

In conclusion, the chemical company should realize that it is wasteful to disregard off-hand
batches that are indicated to have high impurity levels. Further investigation of these batches
should be undertaken to identify the large proportion of them that are in fact satisfactory
products.

1.6.4 Problems

1.6.1 Suppose it is known that 1% of the population suffers from tagged. Species 3 accounts for 17% of the birds, of which
a particular disease. A blood test has a 97% chance of 50% have been tagged. If a tagged bird is observed, what
identifying the disease for diseased individuals, but also are the probabilities that it is of species 1, of species 2, and
has a 6% chance of falsely indicating that a healthy person of species 3?
has the disease.
1.6.5 After production, an electrical circuit is given a quality
(a) What is the probability that a person will have a
score of A, B, C, or D. Over a certain period of time,
positive blood test?
77% of the circuits were given a quality score A,
(b) If your blood test is positive, what is the chance that
11% were given a quality score B, 7% were given a
you have the disease?
quality score C, and 5% were given a quality score D.
(c) If your blood test is negative, what is the chance that
Furthermore, it was found that 2% of the circuits given a
you do not have the disease?
quality score A eventually failed, and the failure rate was
1.6.2 Bag A contains 3 red balls and 7 blue balls. Bag B contains 10% for circuits given a quality score B, 14% for circuits
8 red balls and 4 blue balls. Bag C contains 5 red balls and given a quality score C, and 25% for circuits given a
11 blue balls. A bag is chosen at random, with each bag quality score D.
being equally likely to be chosen, and then a ball is chosen (a) If a circuit failed, what is the probability that it had
at random from that bag. Calculate the probabilities: received a quality score either C or D?
(a) A red ball is chosen. (b) If a circuit did not fail, what is the probability that it
(b) A blue ball is chosen. had received a quality score A?
(c) A red ball from bag B is chosen.
1.6.6 The weather on a particular day is classified as either cold,
If it is known that a red ball is chosen, what is the probability
warm, or hot. There is a probability of 0.15 that it is cold
that it comes from bag A? If it is known that a blue ball
and a probability of 0.25 that it is warm. In addition, on
is chosen, what is the probability that it comes from bag B?
each day it may either rain or not rain. On cold days there is
1.6.3 A class had two sections. Section I had 55 students of a probability of 0.30 that it will rain, on warm days there is
whom 10 received A grades. Section II had 45 students a probability of 0.40 that it will rain, and on hot days there
of whom 11 received A grades. Now 1 of the 100 students is a probability of 0.50 that it will rain. If it is not raining
is chosen at random, with each being equally likely to be on a particular day, what is the probability that it is cold?
chosen.
1.6.7 A valve can be used at four temperature levels. If the valve
(a) What is the probability that the student was in
is used at a cold temperature, then there is a probability of
section I?
0.003 that it will leak. If the valve is used at a medium
(b) What is the probability that the student received an
temperature, then there is a probability of 0.009 that it will
A grade?
leak. If the valve is used at a warm temperature, then there
(c) What is the probability that the student received an
is a probability of 0.014 that it will leak. If the valve is used
A grade if the student is known to have been in
at a hot temperature, then there is a probability of 0.018 that
section I?
it will leak. Under standard operating conditions, the valve
(d) What is the probability that the student was in section I
is used at a cold temperature 12% of the time, at a medium
if the student is known to have received an A grade?
temperature 55% of the time, at a warm temperature 20%
1.6.4 An island has three species of bird. Species 1 accounts for of the time, and at a hot temperature 13% of the time.
45% of the birds, of which 10% have been tagged. Species (a) If the valve leaks, what is the probability that it is
2 accounts for 38% of the birds, of which 15% have been being used at the hot temperature?
56 CHAPTER 1 PROBABILITY THEORY

(b) If the valve does not leak, what is the probability that it there is a probability of 0.60 that the CEO receives a
is being used at the medium temperature? bonus; is slightly above expectation with probability 0.36,
in which case there is a probability of 0.77 that the CEO
1.6.8 A company sells five types of wheelchairs, with type A
receives a bonus; and is considerably above expectation
being 12% of the sales, type B being 34% of the sales,
with probability 0.11, in which case there is a probability
type C being 7% of the sales, type D being 25% of the
of 0.99 that the CEO receives a bonus. What is the
sales, and type E being 22% of the sales. In addition,
probability that the CEO receives a bonus? If the CEO
19% of the type A wheelchair sales are motorized, 50% of
receives a bonus, what is the probability that company has
the type B wheelchair sales are motorized, 4% of the
a revenue below expectation?
type C wheelchair sales are motorized, 32% of the type D
wheelchair sales are motorized, and 76% of the type E 1.6.10 An advertising campaign is canceled before launch with
wheelchair sales are motorized. probability 0.10, in which case the marketing company is
(a) If a motorized wheelchair is sold, what is the fired with probability 0.74; is launched but canceled early
probability that it is of type C? with probability 0.18, in which case the marketing
(b) If a nonmotorized wheelchair is sold, what is the company is fired with probability 0.43; is launched and
probability that it is of type D? runs its targeted length with probability 0.43, in which case
the marketing company is fired with probability 0.16; and
1.6.9 A company’s revenue is considerably below expectation is launched and is extended beyond its targeted length with
with probability 0.08, in which case there is a probability probability 0.29, in which case the marketing company is
of 0.03 that the CEO receives a bonus; is slightly below fired with probability 0.01. What is the probability that the
expectation with probability 0.19, in which case there is a marketing company is fired? If the marketing company is
probability of 0.14 that the CEO receives a bonus; exactly fired, what is the probability that the advertising campaign
meets expectation with probability 0.26, in which case was not canceled before launch?

1.7 Counting Techniques

In many situations the sample space S consists of a very large number of outcomes that
the experimenter will not want to list in their entirety. However, if the outcomes are equally
likely, then it suffices to know the number of outcomes in the sample space and the number of
outcomes contained within an event of interest. In this section, various counting techniques
are discussed that can be used to facilitate such computations. Remember that if a sample
space S consists of N equally likely outcomes, of which n are contained within the event A,
then the probability of the event A is
n
P(A) =
N

1.7.1 Multiplication Rule

Suppose that an experiment consists of k “components” and that the ith component has n i
possible outcomes. The total number of experimental outcomes will then be equal to the
product
n1 × n2 × · · · × nk
This is known as the multiplication rule and can easily be seen by referring to the tree diagram
in Figure 1.70. The n 1 outcomes of the first component are represented by the n 1 branches at
the beginning of the tree, each of which splits into n 2 branches corresponding to the outcomes
of the second component, and so on. The total number of experimental outcomes (the size of
the sample space) is equal to the number of branch ends at the end of the tree, which is equal
to the product of the n i .
 1.7 COUNTING TECHNIQUES 63

so that the number of hands consisting of two kings and two queens is
   
4 4
× = 36
2 2
The required probability is thus
36
 1.33 × 10−4
270,725
which is a chance of about 13 out of 100,000.

1.7.3 Problems

1.7.1 Evaluate: (c) What is the probability that the sample contains no
(a) 7! (b) 8! (c) 4! (d) 13! broken lightbulbs?
(d) How many samples contain exactly 1 broken
1.7.2 Evaluate:
lightbulb?
(a) P27 (b) P59 (c) P25 (d) P417
(e) What is the probability that the sample contains no
1.7.3 Evaluate: more than 1 broken lightbulb?
(a) C26 (b) C48 (c) C25 (d) C614
1.7.9 Show that Ckn = Ckn−1 + Ck−1 n−1
. Can you provide an
1.7.4 A menu has five appetizers, three soups, seven main interpretation of this equality?
courses, six salad dressings, and eight desserts. In how
many ways can a full meal be chosen? In how many ways 1.7.10 A poker hand consists of five cards chosen at random
can a meal be chosen if either an appetizer or a soup is from a pack of cards.
ordered, but not both? (a) How many different hands are there?
(b) How many hands consist of all hearts?
1.7.5 In an experiment to test iron strengths, three different ores, (c) How many hands consist of cards all from the same
four different furnace temperatures, and two different suit (a “flush”)?
cooling methods are to be considered. Altogether, (d) What is the probability of being dealt a flush?
how many experimental configurations are possible? (e) How many hands contain all four aces?
1.7.6 Four players compete in a tournament and are ranked (f) How many hands contain four cards of the same
from 1 to 4. They then compete in another tournament number or picture?
and are again ranked from 1 to 4. Suppose that their (g) What is the probability of being dealt a hand
performances in the second tournament are unrelated to containing four cards of the same number or picture?
their performances in the first tournament, so that the two 1.7.11 In an arrangement of n objects in a circle, an object’s
sets of rankings are independent. neighbors are important, but an object’s place in the circle
(a) What is the probability that each competitor receives is not important. Thus, rotations of a given arrangement
an identical ranking in the two tournaments? are considered to be the same arrangement. Explain why
(b) What is the probability that nobody receives the same the number of different arrangements is (n − 1)!
ranking twice?
1.7.12 In how many ways can six people sit in six seats in a line
1.7.7 Twenty players compete in a tournament. In how at a cinema? In how many ways can the six people sit
many ways can rankings be assigned to the top five around a dinner table eating pizza after the movie?
competitors? In how many ways can the best five
competitors be chosen (without being in any order)? 1.7.13 Repeat Problem 1.7.12 with the condition that one of the
six people, Andrea, must sit next to Scott. In how many
1.7.8 There are 17 broken lightbulbs in a box of 100 lightbulbs.
ways can the seating arrangements be made if Andrea
A random sample of 3 lightbulbs is chosen without
refuses to sit next to Scott?
replacement.
(a) How many ways are there to choose the sample? 1.7.14 A total of n balls are to be put into k boxes with the
(b) How many samples contain no broken lightbulbs? conditions that there will be n 1 balls in box 1, n 2 balls in
64 CHAPTER 1 PROBABILITY THEORY

box 2, and so on, with n k balls being placed in box k he decides that five of the visits will be made one week,
(n 1 + · · · + n k = n). Explain why the number of ways of and the other five visits will be made the following week,
doing this is in how many different ways can the ten cities be split into
n! two groups of five cities?
n1! × · · · × nk !
1.7.20 Suppose that 5 cards are taken without replacement from
Explain why this is just Cnn1 = Cnn2 when k = 2. a deck of 52 cards. How many ways are there to do this so
1.7.15 Explain why the following two problems are identical and that there are 2 red cards and 3 black cards?
solve them.
1.7.21 A hand of 8 cards is chosen at random from an ordinary
(a) In how many ways can 12 balls be placed in 3 boxes,
deck of 52 playing cards without replacement.
when the first box can hold 3 balls, the second box
(a) What is the probability that the hand does not have
can hold 4 balls, and the third box can hold 5 balls.
any hearts?
(b) In how many ways can 3 red balls, 4 blue balls, and 5
(b) What is the probability that the hand consists of two
green balls be placed in a straight line?
(See Problem 1.7.14.) hearts, two diamonds, two clubs, and two spades?

1.7.16 A garage employs 14 mechanics, of whom 3 are needed 1.7.22 A box contains 40 batteries, 5 of which have low
on one job and, at the same time, 4 are needed on another lifetimes, 30 of which have average lifetimes, and 5 of
job. The remaining 7 are to be kept in reserve. In how which have high lifetimes. A consumer requires 8
many ways can the job assignments be made? batteries to run an appliance and randomly selects them
(See Problem 1.7.14.) all from the box. What is the probability that among the
8 batteries fitted into the consumer’s appliance, there are
1.7.17 A company has 15 applicants to interview, and 3 are to be exactly 2 low, 4 average and 2 high lifetimes batteries?
invited on each day of the working week. In how many
ways can the applicants be scheduled? 1.7.23 In each of 3 years a company’s revenue is classified as
(See Problem 1.7.14.) being either considerably below expectation, slightly
below expectation, exactly meeting expectation, slightly
1.7.18 A quality inspector selects a sample of 12 items at
above expectation, or considerably above expectation.
random from a collection of 60 items, of which 18 have
How many different sequences of revenue results are
excellent quality, 25 have good quality, 12 have poor
possible?
quality, and 5 are defective.
(a) What is the probability that the sample only contains 1.7.24 A marketing company is hired to manage four advertising
items that have either excellent or good quality? campaigns sequentially, one after the other. Each
(b) What is the probability that the sample contains three advertising campaign may be canceled before launch,
items of excellent quality, three items of good quality, launched but canceled early, launched and meets its
three items of poor quality, and three defective items? targeted length, or launched and extended beyond its
1.7.19 A salesman has to visit ten different cities. In how many targeted length. How many different sequences of
different ways can the ordering of the visits be made? If campaign results are possible?

1.8 Case Study: Microelectronic Solder Joints

Suppose that using a particular production method there is a probability of 0.85 that a solder
joint has a barrel shape, there is a probability of 0.03 that a solder joint has a cylinder shape,
and there is a probability of 0.12 that a solder joint has an hourglass shape. If it is known
that a particular solder joint does not have a barrel shape, what is the probability that it has a
cylinder shape? This is a conditional probability that can be calculated as
P(cylinder and not barrel) P(cylinder)
P(cylinder|not barrel) = =
P(not barrel) P(cylinder) + P(hourglass)
0.03
= = 0.2
0.03 + 0.12
80 CHAPTER 2 RANDOM VARIABLES

FIGURE 2.17 F(x)

Cumulative distribution function
for dice example 1
17/18
15/18

12/18

8/18

3/18

x
0 1 2 3 4 5

GAMES OF CHANCE The cumulative distribution function for the positive difference between the scores obtained
from two dice can be inferred from the probability mass function given in Figure 2.10 and is
graphed in Figure 2.17. For example, the probability that the difference is no larger than 2 is
1 5 2 2
F(2) = P(X ≤ 2) = P(X = 0) + P(X = 1) + P(X = 2) = + + =
6 18 9 3

2.1.4 Problems

2.1.1 An office has four copying machines, and the random F(x)
variable X measures how many of them are in use
−∞ ≤ x < −4 0.00
at a particular moment in time. Suppose that
P(X = 0) = 0.08, P(X = 1) = 0.11, P(X = 2) = 0.27, −4 ≤ x < −1 0.21
and P(X = 3) = 0.33. −1 ≤ x < 0 0.32
(a) What is P(X = 4)? 0≤x <2 0.39
(b) Draw a line graph of the probability mass function.
2≤x <3 0.68
(c) Construct and plot the cumulative distribution function.
(This problem is continued in Problems 2.3.1 and 2.4.2.) 3≤x <7 0.81
7≤x <∞ 1.00
2.1.2 Figure 2.18 presents the cumulative distribution function
of a random variable. Make a table and line graph of its FIGURE 2.18
probability mass function.
Cumulative distribution function of a random variable
2.1.3 Suppose that two fair dice are rolled and that the two
numbers recorded are multiplied to obtain a final score.
Construct and plot the probability mass function and the (c) What is the most likely value of the random
cumulative distribution function of the final score. (This variable X ?
problem is continued in Problem 2.3.2.) Repeat parts (a)–(c) when the second drawing is made
without replacement. (This problem is continued in
2.1.4 Two cards are drawn at random from a pack of cards with
Problem 2.3.3.)
replacement. Let the random variable X be the number of
cards drawn from the heart suit. 2.1.5 Two fair dice, one red and one blue, are rolled. A score is
(a) Construct the probability mass function. calculated to be twice the value of the blue die if the red
(b) Construct the cumulative distribution function. die has an even value, and to be the value of the red die
 2.2 CONTINUOUS RANDOM VARIABLES 81

minus the value of the blue die if the red die has an odd 2.1.8 Four cards are labeled $1, $2, $3, and $6. A player pays $4,
value. Construct and plot the probability mass function and selects two cards at random, and then receives the sum of
the cumulative distribution function of the score. the winnings indicated on the two cards. Calculate the
probability mass function and the cumulative distribution
2.1.6 A fair coin is tossed three times. A player wins $1 if the
function of the net winnings (that is, winnings minus the
first toss is a head, but loses $1 if the first toss is a tail.
$4 payment).
Similarly, the player wins $2 if the second toss is a head,
but loses $2 if the second toss is a tail, and wins or 2.1.9 A company has five warehouses, only two of which have
loses $3 according to the result of the third toss. Let the a particular product in stock. A salesperson calls the five
random variable X be the total winnings after the three warehouses in a random order until a warehouse with the
tosses (possibly a negative value if losses are incurred). product is reached. Let the random variable X be the
(a) Construct the probability mass function. number of calls made by the salesperson, and calculate its
(b) Construct the cumulative distribution function. probability mass function and cumulative distribution
(c) What is the most likely value of the random function. (This problem is continued in Problems 2.3.4
variable X ? and 2.4.3.)

2.1.7 Consider Example 5 and the probability values given in 2.1.10 Suppose that a random variable X can take the value 1, 2,
Figure 1.38. The company has decided that each television or any other positive integer. Is it possible that
set should be given a quality score calculated in the c
following manner. A perfect picture scores 4, a good P(X = i) =
i2
picture scores 2, a satisfactory picture scores 1, and a failed
picture scores 0. Also, a perfect appearance scores 3, a for some value of the constant c? Is it possible that
good appearance scores 2, a satisfactory appearance scores c
P(X = i) =
1, and a failed appearance scores 0. The final quality score i
is obtained by multiplying the picture score and the for some value of c?
appearance score. For example, the outcome (P, G),
where the picture is perfect and the appearance is good, 2.1.11 A consultant has six appointment times that are open,
would be assigned a quality score of 4 × 2 = 8. three on Monday and three on Tuesday. Suppose that when
(a) Construct the probability mass function of the quality making an appointment a client randomly chooses one of
score. the remaining open times, with each of those open times
(b) Construct the cumulative distribution function of the equally likely to be chosen. Let the random variable X be
quality score. the total number of appointments that have already been
(c) What is the most likely value of the quality score? made over both days at the moment when Monday’s
Recall that an appliance is not shipped if it fails on either schedule has just been completely filled.
the picture or the appearance evaluation, or if it is (a) What is the state space of the random variable X ?
evaluated as only satisfactory in both cases. How can the (b) Calculate the probability mass function and the
probability that an appliance is not shipped be obtained cumulative distribution function of X .
from the cumulative distribution function? (This problem is continued in Problems 2.3.16 and 2.4.12.)

2.2 Continuous Random Variables

2.2.1 Examples of Continuous Random Variables

Random variables are classified as either discrete or continuous according to the set of values
that they can take. Continuous random variables can take any value within a continuous region,
as illustrated in the following examples.

Example 14 A company manufactures metal cylinders that are used in the construction of a particular type
Metal Cylinder of engine. These cylinders, which must slide freely within an outer casing, are designed to
Production have a diameter of 50 mm. The company discovers, however, that the cylinders it manufactures
can have a diameter anywhere between 49.5 and 50.5 mm.
 2.2 CONTINUOUS RANDOM VARIABLES 91

x
F(x) =
1000

P(250 ≤ X ≤ 750) = 1/2

P(X ≤ 750) = 3/4

45°
135°
P(X ≤ 250) = 1/4

0 250 750 1000

Winnings ($)
FIGURE 2.30 FIGURE 2.31

Cumulative distribution function for dial-spinning game illustrating Dial region corresponding to winnings
P(250 ≤ X ≤ 750) between $250 and $750

2.2.4 Problems

15 x
2.2.1 Consider a random variable measuring the following f (x) = + , −2 ≤ x ≤ 0
quantities. In each case state with reasons whether you 64 64
3
think it more appropriate to define the random variable as f (x) = + cx, 0≤x ≤3
8
discrete or as continuous.
and f (x) = 0 elsewhere.
(a) A person’s height
(a) Find the value of c and sketch the probability density
(b) A student’s course grade
function.
(c) The thickness of a metal plate
(b) What is P(−1 ≤ X ≤ 1)?
(d) The purity of a chemical solution
(c) Construct and sketch the cumulative distribution
(e) The type of personal computer a person owns
function.
(f) A person’s age
2.2.4 A random variable X takes values between 0 and 4 with a
2.2.2 A random variable X takes values between 4 and 6 with a
cumulative distribution function
probability density function
x2
1 F(x) =
f (x) = 16
x ln(1.5) for 0 ≤ x ≤ 4.
for 4 ≤ x ≤ 6 and f (x) = 0 elsewhere. (a) Sketch the cumulative distribution function.
(a) Make a sketch of the probability density function. (b) What is P(X ≤ 2)?
(b) Check that the total area under the probability density (c) What is P(1 ≤ X ≤ 3)?
function is equal to 1. (d) Construct and sketch the probability density function.
(c) What is P(4.5 ≤ X ≤ 5.5)? (This problem is continued in Problems 2.3.11 and 2.4.6.)
(d) Construct and sketch the cumulative distribution
2.2.5 A random variable X takes values between 0 and ∞ with
function.
a cumulative distribution function
(This problem is continued in Problems 2.3.10 and 2.4.5.)
F(x) = A + Be−x
2.2.3 A random variable X takes values between −2 and 3 with
a probability density function for 0 ≤ x ≤ ∞.
92 CHAPTER 2 RANDOM VARIABLES

(a) Find the values of A and B and sketch the cumulative 2.2.9 An archer shoots an arrow at a circular target with a
distribution function. radius of 50 cm. If the arrow hits the target, the distance r
(b) What is P(2 ≤ X ≤ 3)? between the point of impact and the center of the target is
(c) Construct and sketch the probability density function. measured. Suppose that this distance has a cumulative
distribution function
2.2.6 A car panel is spray-painted by a machine, and the B
technicians are particularly interested in the thickness of F(r ) = A +
(r + 5)3
the resulting paint layer. Suppose that the random
for 0 ≤ r ≤ 50.
variable X measures the thickness of the paint in
(a) Find the values of A and B and sketch the cumulative
millimeters at a randomly chosen point on a randomly
distribution function.
chosen car panel, and that X takes values between 0.125
(b) What is the probability that the arrow hits within
and 0.5 mm with a probability density function of
10 cm of the center of the target?
f (x) = A(0.5 − (x − 0.25)2 ) (c) What is the probability that the arrow hits more than
for 0.125 ≤ x ≤ 0.5 and f (x) = 0 elsewhere. 30 cm away from the center of the target?
(a) Find the value of A and sketch the probability density (d) Construct and sketch the probability density function.
function. (This problem is continued in Problems 2.3.14 and 2.4.9.)
(b) Construct and sketch the cumulative distribution 2.2.10 Sometimes a random variable is a mix of discrete and
function. continuous components. For example, suppose that the
(c) What is the probability that the paint thickness at a dial-spinning game is modified in the following way. First
particular point is less than 0.2 mm? a fair coin is tossed and if a head is obtained, the player
(This problem is continued in Problems 2.3.12 and 2.4.7.) wins $500 and the dial is not spun. However, if a tail is
2.2.7 Suppose that the random variable X is the time taken by a obtained, the player spins the dial and receives winnings of
garage to service a car. These times are distributed between θ
$1000 ×
0 and 10 hours with a cumulative distribution function 180
as before. In this game there is a probability of 0.5 of
F(x) = A + B ln(3x + 2)
winning $500, with all the other possible winnings
for 0 ≤ x ≤ 10. between $0 and $1000 being equally likely. The coin toss
(a) Find the values of A and B and sketch the cumulative provides a discrete element to the winnings, and the dial
distribution function. spin provides a continuous element. The best way to
(b) What is the probability that a repair job takes longer describe the probabilistic properties of mixed random
than two hours? variables such as this is through a cumulative distribution
(c) Construct and sketch the probability density function. function. The cumulative distribution function of the
winnings from this game is given in Figure 2.32.
2.2.8 The bending capabilities of plastic sheets are investigated
(a) What is the probability of winning less than $200?
by bending sheets at increasingly large angles until a
(b) What is the probability of winning between $400 and
deformity appears in the sheet. The angle θ at which the
$700?
deformity first appears is then recorded. Suppose that this
Interpret your answers.
angle takes values between 0◦ and 10◦ with a probability
density function 2.2.11 The resistance X of an electrical component has a
f (θ) = A(e10−θ − 1) probability density function
f (x) = Ax(130 − x 2 )
for 0 ≤ θ ≤ 10 and f (θ) = 0 elsewhere.
(a) Find the value of A and sketch the probability density for resistance values in the range 10 ≤ x ≤ 11.
function. (a) Calculate the value of the constant A.
(b) Construct and sketch the cumulative distribution (b) Calculate the cumulative distribution function.
function. (c) What is the probability that the electrical component
(c) What is the probability that a plastic sheet can be bent has a resistance between 10.25 and 10.5?
up to an angle of 8◦ without deforming? (This problem is continued in Problems 2.3.17, 2.4.13,
(This problem is continued in Problems 2.3.13 and 2.4.8.) and 2.6.12.)
 2.3 THE EXPECTATION OF A RANDOM VARIABLE 101

FIGURE 2.43 f(x) = 0.128e −x/100

Comparison of median and mean F(133.9) = 0.5 E(X) = 134.1
concrete breaking strengths

x
120 150
Breaking strength

This example illustrates that the median and the mean can both provide useful information
about a random variable. If an engineer has one appliance and wants to know how many
batteries will be required to keep it running over a long period of time, then the mean value
of one hour provides the most useful information. To keep the appliance running for one day,
the engineer can plan on needing about 24 batteries on average.
On the other hand, if the engineer has many appliances and wishes to know how many
of them will last more than 30 minutes on newly charged batteries, then a median value of
about 25 minutes indicates that the engineer can expect fewer than half of the appliances to
last more than 30 minutes.

Example 16 The median concrete breaking strength is the solution to

Concrete Slab Breaking F(x) = 3.856 − 12.8e−x/100 = 0.5
Strengths
which is x = 133.9. Figure 2.43 illustrates this together with the expected breaking strength
of 134.1. The median and the mean strengths are about the same here since the probability
density function is almost flat.

2.3.4 Problems

2.3.1 Consider again the four copying machines discussed in 2.3.4 Consider again the salesperson discussed in Problem 2.1.9
Problem 2.1.1. What is the expected number of copying who is trying to locate a particular product. What is the
machines in use at a particular moment in time? expected number of warehouses called by the salesperson?

2.3.2 Consider again Problem 2.1.3 where the numbers 2.3.5 Suppose that a player draws a card at random from a pack
obtained on two fair dice are multiplied to obtain a final of cards, and wins $15 if an ace, king, queen, or jack is
score. What is the expected value of this score? obtained, and otherwise wins the face value of the card in
dollars. What is the expected amount won by the player?
2.3.3 Consider again Problem 2.1.4 where two cards are drawn
Would you pay $9 to play this game?
from a pack of cards. Is the expected number of hearts
drawn larger when the second drawing is made with or 2.3.6 Two fair dice, one red and one blue, are rolled, and a fair
without replacement? Does this answer surprise you? coin is tossed. If a head is obtained on the coin toss, then
102 CHAPTER 2 RANDOM VARIABLES

a player wins the sum of the scores on the two dice. If a (a) What is the expected value of this random variable?
tail is obtained on the coin toss, then the player wins (b) What is the median of this random variable?
the score on the red die. What are the expected
2.3.12 Consider again the car panel painting machine discussed
winnings?
in Problem 2.2.6. What is the expected paint thickness?
2.3.7 A player pays $1 to play a game where three fair dice are What is the median paint thickness?
rolled. If three 6s are obtained the player wins $500, and
2.3.13 Consider again the plastic bending capabilities discussed
otherwise the player wins nothing. What are the expected
in Problem 2.2.8. What is the expected deformity angle?
net winnings of this game? Would you want to play this
What is the median deformity angle?
game? Does your answer depend upon how many times
you can play the game? 2.3.14 Consider again the archery problem discussed in
Problem 2.2.9. What is the expected deviation from the
2.3.8 A state lottery generally consists of many tickets being
center of the target? What is the median deviation?
sold at prices of about $1, each with a chance to win a
large jackpot which is often over $1 million. Would you 2.3.15 Prove that a continuous random variable with a
expect the expected net winnings on each ticket to be probability density function that is symmetric about a
positive or negative? Why do people play lotteries? point μ has an expected value equal to the point of
symmetry μ.
2.3.9 Suppose that you are organizing the game described at
the end of Section 2.3.1, where you charge players $2 to 2.3.16 Recall Problem 2.1.11 concerning the scheduling of
roll two dice, and then you pay them the difference in the appointments with a consultant. What is the expected
scores. If you fix the dice so that each die has a probability value of the total number of appointments that have
of 0.2 of scoring a 3 and equal probabilities of 0.16 of already been made over both days at the moment when
scoring the other five numbers, do your expected winnings Monday’s schedule has just been completely filled?
increase beyond 6 cents per game? Is this a surprise?
2.3.17 Recall Problem 2.2.11 concerning the resistance of an
2.3.10 Consider again the random variable described in electrical component.
Problem 2.2.2 with a probability density function of (a) What is the expected value of the resistance?
1 (b) What is the median value of the resistance?
f (x) =
x ln(1.5) 2.3.18 A random variable has a probability density function
for 4 ≤ x ≤ 6 and f (x) = 0 elsewhere. f (x) = A(x − 1.5) over the state space 2 ≤ x ≤ 3.
(a) What is the expected value of this random variable? (a) What is the value of A?
(b) What is the median of this random variable? (b) What is the median of the random variable?
2.3.11 Consider again the random variable described in 2.3.19 A manager notices that the number of items purchased by
Problem 2.2.4 with a cumulative distribution function of visitors to a store is zero with probability 0.38, one with
x2 probability 0.44, two with probability 0.15, and three with
F(x) =
16 probability 0.03. What is the expected number of items
for 0 ≤ x ≤ 4. purchased by a visitor to the store?

2.4 The Variance of a Random Variable

2.4.1 Definition and Interpretation of Variance

Another important summary measure of the distribution of a random variable is the variance,
which measures the spread or variability in the values taken by the random variable. Whereas
the mean or expectation measures the central or average value of the random variable, the
variance measures the spread or deviation of the random variable about its mean value.
Specifically, the variance of a random variable is defined as

Var(X ) = E (X − E(X ))2

 2.4 THE VARIANCE OF A RANDOM VARIABLE 113

2.4.5 Problems

2.4.1 Suppose that the random variable X takes the values −2, (c) Find the upper and lower quartiles of this random
1, 4, and 6 with probability values 1/3, 1/6, 1/3, and 1/6, variable.
respectively. (d) What is the interquartile range?
(a) Find the expectation of X .
2.4.7 Consider again the car panel painting machine discussed
(b) Find the variance of X using the formula
in Problems 2.2.6 and 2.3.12.
Var(X ) = E((X − E(X ))2 ) (a) What is the variance of the paint thickness?
(c) Find the variance of X using the formula (b) What is the standard deviation of the paint thickness?
Var(X ) = E(X 2 ) − (E(X ))2 (c) Find the upper and lower quartiles of the paint
thickness.
2.4.2 Consider again the four copying machines discussed in (d) What is the interquartile range?
Problems 2.1.1 and 2.3.1. Calculate the variance and
standard deviation of the number of copying machines in 2.4.8 Consider again the plastic bending capabilities discussed
use at a particular moment. in Problems 2.2.8 and 2.3.13.
(a) What is the variance of the deformity angle?
2.4.3 Consider again the salesperson discussed in Problems (b) What is the standard deviation of the deformity
2.1.9 and 2.3.4 who is trying to locate a particular angle?
product. Calculate the variance and standard deviation of (c) Find the upper and lower quartiles of the deformity
the number of warehouses called by the salesperson. angle.
2.4.4 Suppose that you are organizing the game described at the (d) What is the interquartile range?
end of Section 2.3.1, where you charge players $2 to roll 2.4.9 Consider again the archery problem discussed in
two dice and then you pay them the difference in the Problems 2.2.9 and 2.3.14.
scores. What is the variance in your profit from each (a) What is the variance of the deviation from the center
game? If you are playing a game in which you have of the target?
positive expected winnings, would you prefer a small or a (b) What is the standard deviation of the deviation from
large variance in the winnings? the center of the target?
2.4.5 Consider again the random variable described in Problems (c) Find the upper and lower quartiles of the deviation
2.2.2 and 2.3.10 with a probability density function of from the center of the target.
1 (d) What is the interquartile range?
f (x) =
x ln(1.5) 2.4.10 The time taken to serve a customer at a fast-food
for 4 ≤ x ≤ 6 and f (x) = 0 elsewhere. restaurant has a mean of 75.0 seconds and a standard
(a) What is the variance of this random variable? deviation of 7.3 seconds. Use Chebyshev’s inequality to
(b) What is the standard deviation of this random calculate time intervals that have 75% and 89%
variable? probabilities of containing a particular service time.
(c) Find the upper and lower quartiles of this random
2.4.11 A machine produces iron bars whose lengths have a mean
variable.
of 110.8 cm and a standard deviation of 0.5 cm. Use
(d) What is the interquartile range?
Chebyshev’s inequality to obtain a lower bound on the
2.4.6 Consider again the random variable described in probability that an iron bar chosen at random has a length
Problems 2.2.4 and 2.3.11 with a cumulative distribution between 109.55 cm and 112.05 cm.
function of
2.4.12 Recall Problems 2.1.11 and 2.3.16 concerning the
x2
F(x) = scheduling of appointments with a consultant. What
16 is the standard deviation of the total number of
for 0 ≤ x ≤ 4. appointments that have already been made over both
(a) What is the variance of this random variable? days at the moment when Monday’s schedule has just
(b) What is the standard deviation of this random been completely filled?
variable?
114 CHAPTER 2 RANDOM VARIABLES

2.4.13 Recall Problems 2.2.11 and 2.3.17 concerning the (c) What is the expected value of X ?
resistance of an electrical component. (d) What is the standard deviation of X ?
(a) What is the standard deviation of the resistance? (e) What is the median of X ?
(b) What is the 80th percentile of the resistance? What is (f) What is the upper quartile of X ?
the 10th percentile of the resistance?
2.4.17 When a construction project is opened for bidding,
2.4.14 A continuous random variable has a probability density two proposals are received with probability 0.11, three
function f (x) = Ax 2.5 for 2 ≤ x ≤ 3. proposals are received with probability 0.19, four
(a) What is the value of A? proposals are received with probability 0.55, and
(b) What is the expectation of the random variable? five proposals are received with probability 0.15.
(c) What is the standard deviation of the random (a) What is the expectation of the number of proposals
variable? received?
(d) What is the median of the random variable? (b) What is the standard deviation of the number of
2.4.15 In a game a player either loses $1 with a probability 0.25, proposals received?
wins $1 with a probability 0.4, or wins $4 with a 2.4.18 A random variable X has a distribution given by the
probability 0.35. What are the expectation and the probability density function f (x) = (1 − x)/2 with a
standard deviation of the winnings? state space −1 ≤ x ≤ 1.
2.4.16 A random variable X has a probability density function (a) What is the expected value of X ?
√
f (x) = A/ x for 3 ≤ x ≤ 4. (b) What is the standard deviation of X ?
(a) What is the value of A? (c) What is the upper quartile of X ?
(b) What is the cumulative distribution function of X ?

2.5 Jointly Distributed Random Variables

2.5.1 Joint Probability Distributions

Instead of considering one random variable X and its probability distribution, it is often
appropriate to consider two random variables X and Y and their joint probability distribution.
If the random variables are discrete, then the joint probability mass function consists of
probability values P(X = xi , Y = y j ) = pi j ≥ 0 satisfying

pi j = 1
i j

If the random variables are continuous, then the joint probability density function is a
function f (x, y) ≥ 0 satisfying
 
f (x, y) d x d y = 1
state space

The probability that a ≤ X ≤ b and c ≤ Y ≤ d is obtained from the joint probability density
function as
 b  d
f (x, y) d y d x
x =a y =c

The joint cumulative distribution function is defined to be

F(x, y) = P(X ≤ x, Y ≤ y),
which is
 
F(x, y) = pi j
i:xi ≤x j:y j ≤y