Assignment 8

Reading Assignment:
1. Chapter 8: Continuous Random Variables.
Problems:
1. Let X be a random variable with probability density function
(


C 1 x2
1 < x < 1
f (x) =
0
otherwise.
(a) What is the value of C?
Solution: We use the normalization property to compute the value of C,
Z 1
Z
Z 1




1 2 d
f ()d =
C 1 2 d = C


1
3 
4C
=C
.
=
3 1
3
Thus, we have C = 3/4.
(b) What is the cumulative distribution of X?
Solution: For x [1, 1], we have
Z x
Z x


3
1 2 d
FX (x) =
f ()d =

1 4

x


3

3

x3 2
3
=

+
.
=
4
3 1 4
3
3
When x < 1, FX (x) = 0; similarly, when x > 1, FX (x) = 1.
2. A system consisting of one original unit plus a spare can function for a random amount of
time X. If the density of X is given (in units of months) by
(
Cxex/2 x > 0
f (x) =
0
x 0.
What is the probability that the system functions for at least 5 months?
Solution: First, we find the value of C using the normalization property,
Z
Z
Z
/2
f ()d =
Ce
d = C
e/2 d

0
0


 Z
/2
/2 
= C 2e
2e
d
 +
0
0
 


= C 4e/2 
= 4C.
0

We gather that C = 1/4. The probability that the system functions for at least 5 months is
Z
/2
5
7
Pr(X 5) =
e
d = e5/2 + e5/2 = e5/2 .
4
2
2
5
3. Consider the function
f (x) =

(


C 2x x3
0

0<x<

5
2

otherwise.

Could f be a probability density function? If so, determine C.

Solution: This cannot be a probability density function because it is negative for certain
values between 0 and 5/2. For example, f (1) = C and f (2) = 4C. By definition, the
cumulative distribution function is monotone increasing, and therefore a probability density
function must be non-negative.
Repeat if f (x) were given by
f (x) =

(


C 2x x2
0

0<x<

5
2

otherwise.

Solution: This second candidate suffers from the same problem. In particular, we have
f (1) = C and f (2.5) = 1.25C.
4. The probability density function of X, the lifetime of a certain type of electronic device
(measured in hours), is given by
(
10
x > 10
2
f (x) = x
0
x 10.
(a) Find Pr(X > 20).
Solution: The probability that X exceeds 20 is given by

Z
Z
10
10  1
Pr(X > 20) =
f ()d =
d =  = .
2
20 2
20
20
(b) What is the cumulative distribution function of X?
Solution: For x < 10, we have FX (x) = 0. For x 10, the cumulative distribution
function becomes
Z x
Z x
10
10
d = 1 .
FX (x) =
f ()d =
2

10
(c) What is the probability that, of 6 devices, at least 3 will function for at least 15 hours?
What assumptions are you making?
Solution: The probability that a single specific device will function for at least 15 hours
is 1 FX (15) = 2/3. If failure events are independent across devices, the probability
that at least 3 will function for at least 15 hours is
  3   4   5   6
6 2
6 2
6 2
6 2
+
+
+
.
6
6
6
3 3
4 3
5 3
6 36
2

5. (a) A fire station is to be located along a road of length A, A < . If fires will occur at
points uniformly chosen on (0, A), where should the station be located so as to minimize
the expected distance from the fire? That is, choose a so as to
min E[|X a|]
when X is uniformly distributed over (0, A).
Solution: First, we compute the value of the expectation for a (0, A),
Z A
| a|
E[|X a|] =
d
A
0
Z A
Z a
a
a
d +
d
=
A
A
a
0
a2
(A a)2
=
+
2A
2A
A2 2Aa + 2a2
=
.
2A
To optimize this expression, we differentiate with respect to a and set the derivative
equal to zero,
d A2 2Aa + 2a2
4a 2A
=
= 0.
da
2A
2A
This leads to the solution a = A/2, and a second derivative test ensures that this is a
minimum.
(b) Now suppose that the road is infinite length stretching from point 0 outward to . If
the distance of a fire from point 0 is exponentially distributed with rate , where should
the fire station now be located? That is, we want to minimize E[|X a|], where x is
now exponential with rate .
Solution: The solution is obtained through similar steps. We calculate the value of the
expectation for a (0, ),
Z
E[|X a|] =
| a|e d
0
Z a
Z

=
(a )e d +
( a)e d
a
Z0 a
Z
(a)
=
e
d +
e(+a) d
0
0


a
e
1
ea
a
a
=e
ae
+
+

a
1
e
=a +2
.

To optimize this expectation, we differentiate with respect to a and set the derivative
equal to zero,


d
1
ea
a +2
= 1 2ea = 0.
da

This leads to the solution a = log(2)/, and a second derivative test ensures that this
is a minimum.
6. The time (in hours) required to repair a machine is an exponentially distributed random
variable with parameter = 12 . What is
(a) the probability that a repair time exceeds 2 hours;
Solution: The probability that a repair time exceeds 2 hours is
Z
1 t/2
Pr(T > T2 ) =
e
dt = e1 .
2
2
(b) the conditional probability that a repair takes at least 10 hours, given that its duration
exceeds 9 hours?
Solution: The exponential distribution has the memoryless property. As such, we have
Z
1 t/2
e
dt = e1/2 .
Pr(T > 10|T > 9) = Pr(T > 1) =
2
1
That is, the conditional probability that a repair takes at least 10 hours, given that its
duration exceeds 9 hours, is e1/2 .