Homework 3 Solution

Fall 2021
GU4203/5203

For any errors, comments, or questions,

please email [email protected].

Problem 4.28
4
3 · 20 = 35

Problem 4.29
If check 1, then (if desired) 2: Expected Cost = C1 + (1 − p)C2 + pR1 + (1 − p)R2 ;
if check 2, then 1: Expected Cost = C2 + pC1 + pR1 + (1 − p)R2 . Therefore, checking
p
the second then first is best if C1 + (1 − p)C2 ≤ C2 + pC1 , or C1 ≤ 1−p C2 .

Problem 4.35
If X is the amount that you win, then
4
P (X = 1.10) = = 1 − P (X = −1)
9
4 5 0.6
E[X] = (1.1) − = − ≈ −0.067
9 9 9
 2
24 5 0.6
V ar(X) = (1.1) + − ≈ 1.089
9 9 9

Problem 4.38

E[X 2 ] = [(40)3 + (33)3 + (25)3 + (50)3 ]/148 ≈ 1625.4

V ar(x) = E[X 2 ] − (E[X])2 ≈ 82.2
E[Y 2 ] = [(40)2 + (33)2 + (25)2 + (50)2 ]/4 = 1453.5
V ar(Y ) = 84.5

Problem 4.39
(a) E[(2 + X)2 ] = V ar(2 + X) + (E[2 + X])2 = V ar(X) + 9 = 14
(b) V ar(4 + 3X) = 9V ar(X) = 45

Problem 4.42

10    10
X 10 1
i=7
i 2

1
Exercise 4.28
P (X = n + k)
P (X = n + k|X > n) =
P (X > n)
p(1 − p)n+k−1
=
(1 − p)n
= p(1 − p)k−1

If the first n trials are fall failures, then it is as if we are beginning anew at that time.
Exercise 4.30
Np

N −np

P (X = k + 1) k+1 n−k−1 (N p − k)(n − k)
= N p
N −N p
=
P (X = k) k n−k
(k + 1)(N − N p − n + k + 1)

Problem
Z 1 5.1
3
(a) c (1 − x2 )dx = 1, c = .
−1 4
3 x x3
Z
3 2
(b) F (x) = (1 − x2 )dx = (x − + ), −1 < x < 1.
4 −1 4 3 3

Problem
Z ∞ 5.4  ∞
10 −10 1
(a) 2
dx = = .
20 x Z x 20 2
y
10
(b) F (y) = = 1 − , y > 10. F (y) = 0 for y < 10.
10 y
   i  6−i
6 2 1 10
(c) sum6i=3 since F̄ (15) = . Assuming independence of the events
i 3 3 15
that the devices exceed 15 hours.

Problem 5.11
X is uniform on (0, L).
       
X L−X 1 X L−X 1
P min , < = 1 − P min , >
L−X X 4 L−X X 4
 
X 1 L−X 1
=1−P > , >
L−X 4 X 4
 
L 4L
= 1 − P X > ,X <
5 5
 
L 4L
=1−P <X<
5 5
3 2
=1− =
5 5
Problem 5.15
(a) Φ(0.8333) = 0.7977
(b) 2Φ(1) − 1 = 0.6827
(c) 1 − Φ(0.3333) = 0.3695
(d) Φ(1.6667) = 0.9522
(e) 1 − Φ(1) = 0.1587

Problem 5.19
Let Z = (X − 12)/2. Then, Z is a standard normal. Now, 0.10 = P (Z > (c − 12)/2).

2
But from Table 5.1, P (Z < 1.28) = 0.90, so (c − 12)/2 = 1.28, or c = 14.56.

Exercise 5.2

Z ∞ Z ∞ Z −y
P (Y < −y)dy = fY (x)dxdy
0 0 −∞
Z 0 Z −x
= fY (x)dydx
−∞ 0
Z 0
=− xfY (x)dx
−∞

Similarly,
Z ∞ Z ∞
P (Y > y)dy = xfY (x)dx
0 0

Subtracting these equalities gives the result.

Exercise 5.5
Z (
E[X n ] = X n > t)dt
Z0 ∞
= P (X n > xn )nxn−1 dx by t = xn , dt = nxn−1 dx
0
Z ∞
= P (X n > xn )dx
0

Exercise 5.9
The final steps of parts (a) and (b) use that −Z is also a standard normal random
variable.
(a) P (Z > x) = P (−Z < −x) = P (Z < −x)
(b) P (|Z| > x) = P (Z > x) + P (Z < −x) = P (Z > x) + P (−Z > x)
(c) P (|Z| < x) = 1 − P (|Z| > x) = 1 − 2P (Z > z) = 1 − 2(1 − P (Z < x)), where the
second to last equality follows from (b).

Exercise 5.11
(a) By integration by parts, we have
Z ∞
0 1 z2
E[g (x)] = g 0 (x) √ e− 2 dz
−∞ 2π
Z ∞ √
1 − z2 ∞ z2
= g(z) √ e 2 |−∞ − g(z)(−z) 2πe− 2 dz
2π −∞
Z ∞ √ z2
=0+ zg(z) 2πe− 2 dz
−∞

= E[zg(z)]

(b) Let g(z) = z n , g 0 (z) = nz n−1 . Then.

E[g 0 (z)] = E[nz n−1 ] = nE[z n−1 ],

E[zg(z)] = E[z n+1 ].

3
Since E[g 0 (z)] = E[zg(z)] from (a), we have the desired result.
(c) Using (b), we get E[Z 4 ] = 3E[Z 2 ]. Since E[Z 2 ] = V ar(Z) + E 2 [Z] = 1, it follows
that E[Z 4 ] = 3.

Exercise 5.12
Z ∞ Z ∞
2 2
2
E[X ] = P (X > x)2x 2−1
dx = 2 xe−λx dx = E[X] = 2
0 0 λ λ

Exercise 5.13
(a) b+a
2
(b) µ
(c) 1 − e−λm = 1
2 or m = 1/λ · log2

Exercise 5.15
P (cX < x) = P (X < x/c) = 1 − e−λx/c

Exercise 5.16

f (t) 1/a 1
λ(t) = ¯ = = ,0 < t < a
(F )(t) (a − t)/a a−t

Exercise 5.17
If X has distribution function F and density f , then for a > 0

FaX (t) = P (aX ≤ t) = F (t/a)

1
fax = f (t/a)
a
1
a f (t/a) 1
Therefore, λaX (t) = = λX (t/a).
1 − F (t/a) a