224 Chapter 5 Continuous Random Variables

A random variable is said to have a beta distribution with parameters (a, b) if its
probability density function is equal to
1
f (x) = xa−1 (1 − x)b−1 0 … x … 1
B(a, b)
and is equal to 0 otherwise. The constant B(a, b) is given by
! 1
B(a, b) = xa−1 (1 − x)b−1 dx
0

The mean and variance of such a random variable are, respectively,

a ab
E[X] = Var(X) =
a + b (a + b)2 (a+ b + 1)

PROBLEMS
⎧
5.1. Let X be a random variable with probability den- ⎪
⎨ 10
sity function x > 10
f (x) = x2
" ⎪
⎩0 x … 10
c(1 − x2 ) −1 < x < 1
f (x) =
0 otherwise (a) Find P{X > 20}.
(b) What is the cumulative distribution function
(a) What is the value of c? of X?
(b) What is the cumulative distribution function (c) What is the probability that, of 6 such types of
of X? devices, at least 3 will function for at least 15
5.2. A system consisting of one original unit plus a hours? What assumptions are you making?
spare can function for a random amount of time X. 5.5. A filling station is supplied with gasoline once a
If the density of X is given (in units of months) by week. If its weekly volume of sales in thousands of
" gallons is a random variable with probability den-
Cxe−x/2 x > 0 sity function
f (x) =
0 x … 0
"
5(1 − x)4 0 < x < 1
what is the probability that the system functions f (x) =
0 otherwise
for at least 5 months?
5.3. Consider the function what must the capacity of the tank be so that the
" probability of the supply’s being exhausted in a
C(2x − x3 ) 0 < x < 52 given week is .01?
f (x) =
0 otherwise
5.6. Compute E[X]⎧ if X has a density function given by
⎪
⎨ 1 xe−x/2 x > 0
Could f be a probability density function? If so,
determine C. Repeat if f (x) were given by (a) f (x) = 4 ;
⎪
⎩0 otherwise
" "
C(2x − x2 ) 0 < x < 52 c(1 − x2 ) −1 < x < 1
f (x) = (b) f (x) = ;
0 otherwise 0 otherwise
⎧
⎨5 x > 5
⎪
5.4. The probability density function of X, the lifetime (c) f (x) = x2 .
of a certain type of electronic device (measured in ⎪
⎩0 x … 5
hours), is given by
 Problems 225

5.7. The density function of X is given by (b) If, at 10:15, the bus has not yet arrived, what
" is the probability that you will have to wait at
a + bx2 0 … x … 1 least an additional 10 minutes?
f (x) =
0 otherwise 5.14. Let X be a uniform (0, 1) random variable. Com-
pute E[X n ] by using Proposition 2.1, and then
If E[X] = 35 , find a and b. check the result by using the definition of expec-
5.8. The lifetime in hours of an electronic tube is a ran- tation.
dom variable having a probability density function 5.15. If X is a normal random variable with parameters
given by µ = 10 and σ 2 = 36, compute
f (x) = xe−x x Ú 0 (a) P{X > 5};
(b) P{4 < X < 16};
Compute the expected lifetime of such a tube. (c) P{X < 8};
5.9. Consider Example 4 b of Chapter 4 , but now sup- (d) P{X < 20};
pose that the seasonal demand is a continuous ran- (e) P{X > 16}.
dom variable having probability density function f . 5.16. The annual rainfall (in inches) in a certain region is
Show that the optimal amount to stock is the value normally distributed with µ = 4 0 and σ = 4 . What
s∗ that satisfies is the probability that, starting with this year, it will
b take over 10 years before a year occurs having a
F(s∗ ) = rainfall of over 50 inches? What assumptions are
b + ℓ you making?
where b is net profit per unit sale, ℓ is the net loss 5.17. A man aiming at a target receives 10 points if his
per unit unsold, and F is the cumulative distribu- shot is within 1 inch of the target, 5 points if it is
tion function of the seasonal demand. between 1 and 3 inches of the target, and 3 points if
5.10. Trains headed for destination A arrive at the train it is between 3 and 5 inches of the target. Find the
station at 15-minute intervals starting at 7 A . M ., expected number of points scored if the distance
whereas trains headed for destination B arrive at from the shot to the target is uniformly distributed
15-minute intervals starting at 7:05 A . M . between 0 and 10.
(a) If a certain passenger arrives at the station at 5.18. Suppose that X is a normal random variable with
a time uniformly distributed between 7 and mean 5. If P{X > 9} = .2, approximately what is
8 A . M . and then gets on the first train that Var(X)?
arrives, what proportion of time does he or
she go to destination A? 5.19. Let X be a normal random variable with mean
(b) What if the passenger arrives at a time uni- 12 and variance 4 . Find the value of c such that
formly distributed between 7:10 and 8:10 P{X > c} = .10.
A . M .? 5.20. If 65 percent of the population of a large commu-
5.11. A point is chosen at random on a line segment nity is in favor of a proposed rise in school taxes,
of length L. Interpret this statement, and find the approximate the probability that a random sample
probability that the ratio of the shorter to the of 100 people will contain
longer segment is less than 14 . (a) at least 50 who are in favor of the proposition;
(b) between 60 and 70 inclusive who are in favor;
5.12. A bus travels between the two cities A and B,
(c) fewer than 75 in favor.
which are 100 miles apart. If the bus has a break-
down, the distance from the breakdown to city A 5.21. Suppose that the height, in inches, of a 25-year-old
has a uniform distribution over (0, 100). There is a man is a normal random variable with parameters
bus service station in city A, in B, and in the cen- µ = 71 and σ 2 = 6.25. What percentage of 25-
ter of the route between A and B. It is suggested year-old men are over 6 feet, 2 inches tall? What
that it would be more efficient to have the three percentage of men in the 6-footer club are over 6
stations located 25, 50, and 75 miles, respectively, feet, 5 inches?
from A. Do you agree? Why? 5.22. The width of a slot of a duralumin forging is (in
5.13. You arrive at a bus stop at 10 o’clock, knowing inches) normally distributed with µ = .9000 and
that the bus will arrive at some time uniformly dis- σ = .0030. The specification limits were given as
tributed between 10 and 10:30. .9000 ; .0050.
(a) What is the probability that you will have to (a) What percentage of forgings will be defec-
wait longer than 10 minutes? tive?
226 Chapter 5 Continuous Random Variables

(b) What is the maximum allowable value of σ chosen point in the black region will have a nor-
that will permit no more than 1 in 100 defec- mally distributed reading with parameters (6, 9).
tives when the widths are normally distributed A point is randomly chosen on the image and has
with µ = .9000 and σ ? a reading of 5. If the fraction of the image that is
5.23. One thousand independent rolls of a fair die will black is α, for what value of α would the probabil-
be made. Compute an approximation to the prob- ity of making an error be the same, regardless of
ability that the number 6 will appear between 150 whether one concluded that the point was in the
and 200 times inclusively. If the number 6 appears black region or in the white region?
exactly 200 times, find the probability that the 5.31. (a) A fire station is to be located along a road of
number 5 will appear less than 150 times. length A, A < q. If fires occur at points uni-
5.24. The lifetimes of interactive computer chips pro- formly chosen on (0, A), where should the sta-
duced by a certain semiconductor manufacturer tion be located so as to minimize the expected
are normally distributed with parameters µ = distance from the fire? That is, choose a so
1.4 * 106 hours and σ = 3 * 105 hours. What is the as to
approximate probability that a batch of 100 chips minimize E[|X − a|]
will contain at least 20 whose lifetimes are less than
1.8 * 106 ? when X is uniformly distributed over (0, A).
(b) Now suppose that the road is of infinite
5.25. Each item produced by a certain manufacturer is,
length—stretching from point 0 outward to q.
independently, of acceptable quality with proba-
If the distance of a fire from point 0 is expo-
bility .95. Approximate the probability that at most
nentially distributed with rate λ, where should
10 of the next 150 items produced are unaccept-
the fire station now be located? That is, we
able.
want to minimize E[|X − a|], where X is now
5.26. Two types of coins are produced at a factory: a fair exponential with rate λ.
coin and a biased one that comes up heads 55 per-
5.32. The time (in hours) required to repair a machine is
cent of the time. We have one of these coins, but
an exponentially distributed random variable with
do not know whether it is a fair coin or a biased
parameter λ = 12 . What is
one. In order to ascertain which type of coin we
(a) the probability that a repair time exceeds 2
have, we shall perform the following statistical test:
hours?
We shall toss the coin 1000 times. If the coin lands
(b) the conditional probability that a repair takes
on heads 525 or more times, then we shall conclude
at least 10 hours, given that its duration
that it is a biased coin, whereas if it lands on heads
exceeds 9 hours?
less than 525 times, then we shall conclude that it
is a fair coin. If the coin is actually fair, what is the 5.33. The number of years a radio functions is exponen-
probability that we shall reach a false conclusion? tially distributed with parameter λ = 18 . If Jones
What would it be if the coin were biased? buys a used radio, what is the probability that it
will be working after an additional 8 years?
5.27. In 10,000 independent tosses of a coin, the coin
landed on heads 5800 times. Is it reasonable to 5.34. Jones figures that the total number of thousands
assume that the coin is not fair? Explain. of miles that an auto can be driven before it would
need to be junked is an exponential random vari-
5.28. Twelve percent of the population is left handed. 1
able with parameter 20 . Smith has a used car that
Approximate the probability that there are at least
he claims has been driven only 10,000 miles. If
20 left-handers in a school of 200 students. State
Jones purchases the car, what is the probability
your assumptions.
that she would get at least 20,000 additional miles
5.29. A model for the movement of a stock supposes out of it? Repeat under the assumption that the
that if the present price of the stock is s, then, after lifetime mileage of the car is not exponentially dis-
one period, it will be either us with probability p tributed, but rather is (in thousands of miles) uni-
or ds with probability 1 − p. Assuming that suc- formly distributed over (0, 4 0).
cessive movements are independent, approximate
5.35. The lung cancer hazard rate λ(t) of a t-year-old
the probability that the stock’s price will be up
male smoker is such that
at least 30 percent after the next 1000 periods if
u = 1.012, d = 0.990, and p = .52. λ(t) = .027 + .00025(t − 4 0)2 t Ú 40
5.30. An image is partitioned into two regions, one
white and the other black. A reading taken from Assuming that a 4 0-year-old male smoker survives
a randomly chosen point in the white section will all other hazards, what is the probability that he
give a reading that is normally distributed with µ = survives to (a) age 50 and (b) age 60 without con-
4 and σ 2 = 4 , whereas one taken from a randomly tracting lung cancer?
 Theoretical Exercises 227

5.36. Suppose that the life distribution of an item has the 5.39. If X is an exponential random variable with
hazard rate function λ(t) = t3 , t > 0. What is the parameter λ = 1, compute the probability den-
probability that sity function of the random variable Y defined by
(a) the item survives to age 2? Y = log X.
(b) the item’s lifetime is between .4 and 1.4 ? 5.40. If X is uniformly distributed over (0, 1), find the
(c) a 1-year-old item will survive to age 2? density function of Y = eX .
5.41. Find the distribution of R = A sin θ , where A is
5.37. If X is uniformly distributed over (−1, 1), find a fixed constant and θ is uniformly distributed on
(a) P{|X| > 12 }; (−π/2, π/2). Such a random variable R arises in
(b) the density function of the random variable the theory of ballistics. If a projectile is fired from
|X|. the origin at an angle α from the earth with a speed
ν, then the point R at which it returns to the earth
5.38. If Y is uniformly distributed over (0, 5), what can be expressed as R = (v2 /g) sin 2α, where g is
is the probability that the roots of the equation the gravitational constant, equal to 980 centime-
4 x2 + 4 xY + Y + 2 = 0 are both real? ters per second squared.

THEORETICAL EXERCISES

5.1. The speed of a molecule in a uniform gas at equi- 5.5. Use the result that, for a nonnegative random vari-
librium is a random variable whose probability able Y, ! q
density function is given by
E[Y] = P{Y > t} dt
" 0
2 −bx2 x Ú 0
f (x) = ax e to show that, for a nonnegative random vari-
0 x < 0
able X,
where b = m/2kT and k, T, and m denote, respec- ! q
tively, Boltzmann’s constant, the absolute temper- E[X n ] = nxn−1 P{X > x} dx
0
ature of the gas, and the mass of the molecule.
Evaluate a in terms of b. Hint: Start with
5.2. Show that ! q
! q ! q E[X n ] = P{X n > t} dt
0
E[Y] = P{Y > y} dy − P{Y < −y} dy
0 0
and make the change of variables t = xn .
Hint: Show that 5.6. Define a collection of events Ea , 0 < a < 1, hav-
! q ! 0 ing' the property
) that P(Ea ) = 1 for all a but
P{Y < −y} dy = − xfY (x) dx (
P Ea = 0.
0 −q
! q ! q a
P{Y > y} dy = xfY (x) dx Hint: Let X be uniform over (0, 1) and define each
0 0 Ea in terms of X.
5.7. The standard deviation of X, denoted SD(X), is
5.3. Show that if X has density function f , then given by
! q *
SD(X) = Var(X)
E[g(X)] = g(x)f (x) dx
−q Find SD(aX + b) if X has variance σ 2 .
Hint: Using Theoretical Exercise 2, start with 5.8. Let X be a random variable that takes on values
! q ! q between 0 and c. That is, P{0 … X … c} = 1.
Show that
E[g(X)] = P{g(X) > y} dy − P{g(X) < −y} dy
c2
0 0 Var(X) …
4
and then proceed as in the proof given in the text
when g(X) Ú 0. Hint: One approach is to first argue that
5.4. Prove Corollary 2.1.
E[X 2 ] … cE[X]
228 Chapter 5 Continuous Random Variables

and then use this inequality to show that when X is a gamma random variable with param-
eters α and λ.+ ,
E[X] √
Var(X) … c2 [α(1 − α)] where α = 5.21. Show that ( 12 = π.
c + , -q
Hint: ( 12 = 0 e−x x−1/2 dx. Make the change
5.9. Show that Z is a standard normal random variable, √
then, for x > 0, of variables y = 2x and then relate the resulting
(a) P{Z > x} = P{Z < −x}; expression to the normal distribution.
(b) P{|Z| > x} = 2P{Z > x}; 5.22. Compute the hazard rate function of a gamma ran-
(c) P{|Z| < x} = 2P{Z < x} − 1. dom variable with parameters (α, λ) and show it is
5.10. Let f (x) denote the probability density function of increasing when α Ú 1 and decreasing when α … 1.
a normal random variable with mean µ and vari- 5.23. Compute the hazard rate function of a Weibull
ance σ 2 . Show that µ − σ and µ + σ are points random variable and show it is increasing when
of inflection of this function. That is, show that β Ú 1 and decreasing when β … 1.
f ′′ (x) = 0 when x = µ − σ or x = µ + σ . 5.24. Show that a plot of log(log(1 − F(x))−1 ) against
5.11. Let Z be a standard normal random variable Z, log x will be a straight line with slope β when
and let g be a differentiable function with deriva- F(·) is a Weibull distribution function. Show also
tive g′ . that approximately 63.2 percent of all observa-
(a) Show that E[g′ (Z)] = E[Zg(Z)] tions from such a distribution will be less than α.
(b) Show that E[Zn+1 ] = nE[Zn−1 ] Assume that v = 0.
(c) Find E[Z4 ].
5.25. Let
5.12. Use the identity of Theoretical Exercise 5 to derive . /
X − ν β
E[X 2 ] when X is an exponential random variable Y=
with parameter λ. α
5.13. The median of a continuous random variable hav-
Show that if X is a Weibull random variable with
ing distribution function F is that value m such that
parameters ν, α, and β, then Y is an exponential
F(m) = 12 . That is, a random variable is just as random variable with parameter λ = 1 and vice
likely to be larger than its median as it is to be
versa.
smaller. Find the median of X if X is
(a) uniformly distributed over (a, b); 5.26. If X is a beta random variable with parameters a
(b) normal with parameters µ, σ 2 ; and b, show that
(c) exponential with rate λ. a
5.14. The mode of a continuous random variable having E[X] =
a + b
density f is the value of x for which f (x) attains its ab
maximum. Compute the mode of X in cases (a), Var(X) =
(a + b)2 (a + b + 1)
(b), and (c) of Theoretical Exercise 5.13.
5.15. If X is an exponential random variable with 5.27. If X is uniformly distributed over (a, b), what ran-
parameter λ, and c > 0, show that cX is exponen- dom variable, having a linear relation with X, is
tial with parameter λ/c. uniformly distributed over (0, 1)?
5.16. Compute the hazard rate function of X when X is
5.28. Consider the beta distribution with parameters
uniformly distributed over (0, a).
(a, b). Show that
5.17. If X has hazard rate function λX (t), compute the
(a) when a > 1 and b > 1, the density is uni-
hazard rate function of aX where a is a positive
modal (that is, it has a unique mode) with
constant.
mode equal to (a − 1)/(a + b − 2);
5.18. Verify that the gamma density function integrates
(b) when a … 1, b … 1, and a + b < 2, the den-
to 1.
sity is either unimodal with mode at 0 or 1 or
5.19. If X is an exponential random variable with mean U-shaped with modes at both 0 and 1;
1/λ, show that
(c) when a = 1 = b, all points in [0, 1] are modes.
k! 5.29. Let X be a continuous random variable having
E[X k ] = k = 1, 2, . . . cumulative distribution function F. Define the ran-
λk
dom variable Y by Y = F(X). Show that Y is
Hint: Make use of the gamma density function to uniformly distributed over (0, 1).
evaluate the preceding.
5.30. Let X have probability density fX . Find the prob-
5.20. Verify that ability density function of the random variable Y
α
Var(X) = 2 defined by Y = aX + b.
λ
 Self-Test Problems and Exercises 229

q
1
5.31. Find the probability density function of Y = eX
It is a well-known identity that 1/k2 =
when X is normally distributed with parameters µ 1
and σ 2 . The random variable Y is said to have a π 2 /6, so Q1 = 6/π 2 . (In number theory, this
lognormal distribution (since log Y has a normal is known as the Legendre theorem.)
distribution) with parameters µ and σ 2 . (c) Now argue that
5.32. Let X and Y be independent random vari- ' )
2q
ables that are both equally likely to be either Pi2 − 1
1, 2, . . . , (10)N , where N is very large. Let D denote Q1 =
i=1
Pi2
the greatest common divisor of X and Y, and let
Qk = P{D = k}. where Pi is the ith-smallest prime greater
(a) Give a heuristic argument that Qk = k12 Q1 . than 1.
Hint: Note that in order for D to equal k, k Hint: X and Y will be relatively prime if they
must divide both X and Y and also X/k, and have no common prime factors. Hence, from
Y/k must be relatively prime. (That is, X/k, part (b), we see that
and Y/k must have a greatest common divisor ' )
2q
equal to 1.) Pi2 − 1 6
2
= 2
(b) Use part (a) to show that Pi π
i=1

which was noted without explanation in

Q1 = P{X and Y are relatively prime} Problem 11 of Chapter 4 . (The relationship
between this problem and Problem 11 of
1 Chapter 4 is that X and Y are relatively prime
= q
0 if XY has no multiple prime factors.)
1/k2 5.33. Prove Theorem 7.1 when g(x) is a decreasing func-
k=1 tion.

SELF-TEST PROBLEMS AND EXERCISES

5.1. The number of minutes of playing time of a certain 5.3. For some constant c, the random variable X has
high school basketball player in a randomly cho- the probability density function
sen game is a random variable whose probability "
density function is given in the following figure: cx4 0 < x < 2
f (x) =
0 otherwise
.050
Find (a) E[X] and (b) Var(X).
.025 5.4. The random variable X has the probability density
function
"
10 20 30 40 ax + bx2 0 < x < 1
f (x) =
0 otherwise
Find the probability that the player plays
(a) over 15 minutes; If E[X] = .6, find (a) P{X < 12 } and (b) Var(X).
(b) between 20 and 35 minutes; 5.5. The random variable X is said to be a discrete uni-
(c) less than 30 minutes; form random variable on the integers 1, 2, . . . , n if
(d) more than 36 minutes.
5.2. For some constant c, the random variable X has 1
P{X = i} = i = 1, 2, . . . , n
the probability density function n
3 n
cx 0 < x < 1 For any nonnegative real number x, let Int(x)
f (x) = (sometimes written as [x]) be the largest inte-
0 otherwise
ger that is less than or equal to x. Show that
Find (a) c and (b) P{X > x}, 0 < x < 1. U is a uniform random variable on (0, 1), fain
230 Chapter 5 Continuous Random Variables

X = Int(nU) + 1 is a discrete uniform random Assume that if Ai is the event that the rainfall
variable on 1, . . . , n. exceeds 4 4 inches in year i (from now), then the
5.6. Your company must make a sealed bid for a con- events Ai , i Ú 1, are independent.
struction project. If you succeed in winning the 5.12. The following table uses 1992 data concerning the
contract (by having the lowest bid), then you plan percentages of male and female full-time workers
to pay another firm 100 thousand dollars to do whose annual salaries fall into different ranges:
the work. If you believe that the minimum bid
(in thousands of dollars) of the other participating Earnings range Percentage Percentage
companies can be modeled as the value of a ran- of females of males
dom variable that is uniformly distributed on (70,
14 0), how much should you bid to maximize your …9999 8.6 4 .4
expected profit? 10,000–19,999 38.0 21.1
5.7. To be a winner in a certain game, you must be 20,000–24 ,999 19.4 15.8
successful in three successive rounds. The game 25,000–4 9,999 29.2 4 1.5
depends on the value of U, a uniform random vari- Ú50,000 4 .8 17.2
able on (0, 1). If U > .1, then you are successful
in round 1; if U > .2, then you are successful in Suppose that random samples of 200 male and 200
round 2; and if U > .3, then you are successful in female full-time workers are chosen. Approximate
round 3. the probability that
(a) Find the probability that you are successful in (a) at least 70 of the women earn $25,000 or more;
round 1. (b) at most 60 percent of the men earn $25,000
(b) Find the conditional probability that you are or more;
successful in round 2 given that you were suc- (c) at least three-fourths of the men and at least
cessful in round 1. half the women earn $20,000 or more.
(c) Find the conditional probability that you are 5.13. At a certain bank, the amount of time that a cus-
successful in round 3 given that you were suc- tomer spends being served by a teller is an expo-
cessful in rounds 1 and 2. nential random variable with mean 5 minutes. If
(d) Find the probability that you are a winner. there is a customer in service when you enter the
5.8. A randomly chosen IQ test taker obtains a score bank, what is the probability that he or she will still
that is approximately a normal random variable be with the teller after an additional 4 minutes?
with mean 100 and standard deviation 15. What is 5.14. Suppose that the cumulative distribution function
the probability that the score of such a person is of the random variable X is given by
(a) above 125; (b) between 90 and 110?
2
5.9. Suppose that the travel time from your home to F(x) = 1 − e−x x > 0
your office is normally distributed with mean 4 0
minutes and standard deviation 7 minutes. If you Evaluate (a) P{X > 2}; (b) P{1 < X < 3}; (c) the
want to be 95 percent certain that you will not be hazard rate function of F; (d) E[X]; (e) Var(X).
late for an office appointment at 1 P . M ., what is the Hint: For parts (d) and (e), you might want to
latest time that you should leave home? make use of the results of Theoretical Exercise 5.
5.10. The life of a certain type of automobile tire is nor- 5.15. The number of years that a washing machine func-
mally distributed with mean 34 ,000 miles and stan- tions is a random variable whose hazard rate func-
dard deviation 4 000 miles. tion is given by
(a) What is the probability that such a tire lasts ⎧
over 4 0,000 miles? ⎨ .2 0 < t < 2
(b) What is the probability that it lasts between λ(t) = .2 + .3(t − 2) 2 … t < 5
⎩ 1.1 t > 5
30,000 and 35,000 miles?
(c) Given that it has survived 30,000 miles, what
is the conditional probability that the tire sur- (a) What is the probability that the machine will
vives another 10,000 miles? still be working 6 years after being purchased?
5.11. The annual rainfall in Cleveland, Ohio is approx- (b) If it is still working 6 years after being pur-
imately a normal random variable with mean 4 0.2 chased, what is the conditional probability
inches and standard deviation 8.4 inches. What is that it will fail within the next 2 years?
the probability that 5.16. A standard Cauchy random variable has density
(a) next year’s rainfall will exceed 4 4 inches? function
(b) the yearly rainfalls in exactly 3 of the next 7 1
years will exceed 4 4 inches? f (x) = − q < x < q
π(1 + x2 )
 Self-Test Problems and Exercises 231

Show that X is a standard Cauchy random variable X whose mean µ depends on whether the
variable, then 1/X is also a standard Cauchy ran- defendant is guilty. If innocent, µ = 1; if guilty,
dom variable. µ = 2. The deciding judge will rule the defen-
5.17. A roulette wheel has 38 slots, numbered 0, 00, and dant guilty if X > c for some suitably chosen
1 through 36. If you bet 1 on a specified num- value of c.
ber then you either win 35 if the roulette ball (a) If the judge wants to be 95 percent certain
lands on that number or lose 1 if it does not. that an innocent man will not be convicted,
If you continually make such bets, approximate what should be the value of c?
the probability that (b) Using the value of c found in part (a), what is
(a) you are winning after 34 bets; the probability that a guilty defendant will be
(b) you are winning after 1000 bets; convicted?
(c) you are winning after 100,000 bets.
Assume that each roll of the roulette ball is equally 5.20. For any real number y , define y + by
likely to land on any of the 38 numbers.
5.18. There are two types of batteries in a bin. When in y , if y Ú 0
y+ =
use, type i batteries last (in hours) an exponentially 0, if y < 0
distributed time with rate λi , i = 1, 2. A battery
that is randomly chosen from the bin will be a type Let c be a constant.
12 (a) Show that
i battery with probability pi , pi = 1. If a ran-
i=1 1 2
domly chosen battery is still operating after t hours E[(Z − c)+ ] = √ e−c /2 − c(1 − *(c))
of use, what is the probability that it will still be 2π
operating after an additional s hours?
5.19. Evidence concerning the guilt or innocence of a when Z is a standard normal random variable.
defendant in a criminal investigation can be sum- (b) Find E[(X − c)+ ] when X is normal with
marized by the value of an exponential random mean µ and variance σ 2 .