0% found this document useful (0 votes)
112 views5 pages

Solution3 602

This document provides solutions to homework problems involving probability distributions and their properties. It includes calculating the cumulative distribution function, probability density function, expected value, variance, and moment generating function for various random variables. Median values are also determined for some distributions.

Uploaded by

trollycn
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
112 views5 pages

Solution3 602

This document provides solutions to homework problems involving probability distributions and their properties. It includes calculating the cumulative distribution function, probability density function, expected value, variance, and moment generating function for various random variables. Median values are also determined for some distributions.

Uploaded by

trollycn
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Summer 2006 STAT 602 Homework 3 Solution

1. (a) CDF of X is:





 0, x≤0
 R
x
FX (x) = 0
2tdt = x2 , 0<x<1



1, x≥1

1 1
2x3 2
Z
(b) E(X) = x · 2xdx = =3
0 3 0
Z 1 4 1

2 2 x =1
E(X ) = x · 2xdx =
0 2 2 0
2 1
V ar(X) = E(X 2 ) − (E(X)) =
18
(c) For t 6= 0,
1 1 1 Z 1
2x tx 2x tx 2 tx
Z Z
tx
MX (t) = Ee tX
= e · 2xdx = de = e − e dx = 2t−2 (tet − et + 1)
0 0 t t 0 0 t
For t = 0,

MX (t) = EetX = 1
2tet − 2et + 2 2et + 2tet − 2et
and lim 2t−2 (tet − et + 1) = lim 2
= lim = lim et = 1
t→0 t→0 t t→0 2t t→0

Thus, the mgf for X exists.

(d) Let m be the median, by the definition of median,


Z m √
1 2
F (m) = 2xdx = ⇒ m =
0 2 2
Z 1
1 ∞ 1 1
Z ∞ Z
t=− ln x
(e) E(ln X) = ln x · 2xdx = (−t) · 2e−t · (−e−t )dt = − te−2t dt = −
0 0 2 0 1 2
Γ(2) 2
2
2. (a)



 0, x<1


.10, 1≤x<2







.25, 2≤x<3






.35, 3≤x<6


FX (x) =



 .50, 6≤x<7


.65, 7≤x<8






.70, 8 ≤ x < 20







1, x ≥ 20

P
(b) E(X) = xPX (x) = 1(.10) + 2(.15) + 3(.10) + 6(.15) + 7(.15) + 8(.05) + 20(.3) = 9.05

E(X 2 ) = x2 PX (x) = 12 (.10) + 22 (.15) + 32 (.10) + 62 (.15) + 72 (.15) + 82 (.05) + 202 (.3) = 137.55
P

2
V ar(X) = E(X 2 ) − (E(X)) = 55.6475

1
Summer 2006 STAT 602 Homework 3 Solution

(c) The mgf of X is:

MX (t) = EetX = etx PX (x) = .1et + .15e2t + .1e3t + .15e6t + .15e7t + .05e8t + .3e20t
P

dMX (t)
.1et + .3e2t + .3e3t + .9e6t + 1.05e7t + .4e8t + 6e20t t=0

=
dt t=0

= .1 + .3 + .3 + .9 + 1.05 + .4 + 6 = 9.05 = E(X)


2

d MX (t)
.1et + .6e2t + .9e3t + 5.4e6t + 7.35e7t + 3.2e8t + 120e20t t=0

=
dt2 t=0

= .1 + .6 + .9 + 5.4 + 7.35 + 3.2 + 120 = 137.55 = E(X 2 )


2
So V ar(X) = E(X 2 ) − (E(X)) = 55.6475

(d) from (a), FX (x) = 0.5, 6 ≤ x < 7, median m is any value in [6,7).
2 2 2

θeθt (1 − η 2 t2 ) − eθt (−2η 2 t) eθt(θ−θη t +2η t)

dMX (t)
3. E(X) = = = =θ
dt t=0 (1 − η 2 t2 )2 (1 − η 2 t2 )2

t=0 t=0
d2 MX (t)

E(X 2 ) = =
dt2 t=0
{θeθt [θ − θη 2 t2 + 2η 2 t] + eθt [−2θη 2 t + 2η 2 ]}(1 − η 2 t2 )4 + eθt (θ − θη 2 t2 + 2η 2 t)2(1 − η 2 t)(−2η 2 t)

(1 − η 2 t2 )4
t=0
2 2
= θ + 2η
2
V ar(X) = E(X 2 ) − (E(X)) = 2η 2

4. (a) When x > 0,


Z x
1 −t 1

1  −t  x 1 x 
FX (x) = e 2 + te−t dt = − e 2 + te−t + e−t = − e− 2 + xe−x + e−x + 1
0 4 2 2 0 2
When x ≤ 0, FX (x) = 0
1 
(b) P (2 < X < 3) = FX (3) − FX (2) = − e−1.5 + 4e−3 − e−1 − 3e−2 = 0.176
2
  Z ∞ 
1 1 1 (t− 12 )x 1 (t−1)x
Z ∞
x
(c) MX (t) = E(etX ) = etx e− 2 + xe−x dx = e + xe dx
0Z 4 2 0 4 2
1 (t− 21 )x 1 (t−1)x
Z ∞ ∞
= e dx + xe dx
0 4 −1 Z ∞ 0 2
1 1 1 1 1
Z ∞ −x

= −t e( t− 21 )x
dx + (1 − t) −2
xe
1
1−t dx
4 2 1 2 2
Γ(1) 1 −t
 
0 0 1
2 Γ(2) 1−t
 −1
1 1 1 −2 1
= −t + (1 − t) , t <
4 2 2 2
 −2
dMX (t) 1 1
−3
(d) E(X) = = − t + (1 − t) =2
dt t=0 4 2


t=0
d2 MX (t)
 −3
1 1
E(X 2 ) =
−4
= − t + 3 (1 − t) =7
dt2 t=0 2 2


t=0
2 2
V ar(X) = E(X ) − (E(X)) = 3

2
Summer 2006 STAT 602 Homework 3 Solution

5. (a) Rolling two dice, there are totally 36 possible outcomes for (X, Y ). Each has same probability
1/36.

For X < 2Y , the possible outcomes are the following:

X Y
1 1,2,3,4,5,6
2 2,3,4,5,6
3 2,3,4,5,6
4 3,4,5,6
5 3,4,5,6
6 4,5,6

There are total 27 possible outcomes. So P (X < 2Y ) = 27/36 = 0.75

X
Y-X 1 2 3 4 5 6
1 0 -1 -2 -3 -4 -5
2 1 0 -1 -2 -3 -4
(b)
3 2 1 0 -1 -2 -3
Y 4 3 2 1 0 -1 -2
5 4 3 2 1 0 -1
6 5 4 3 2 1 0
From the above table, we know that Y − X can be 5,4,3,2,1,0,-1,-2,-3,-4,-5.

So the probability distribution for Z = Y − X is:

P (Z = −5) = P (Z = 5) = 1/36
P (Z = −4) = P (Z = 4) = 1/18
P (Z = −3) = P (Z = 3) = 1/12
P (Z = −2) = P (Z = 2) = 1/9
P (Z = −1) = P (Z = 1) = 5/36
P (Z = 0) = 1/6

5
P
(c) E(Z) = i · P (Z = i) = 0
i=−5
5
E(Z 2 ) = i2 · P (Z = i) = 35/6
P
i=−5
V ar(Z) = 35/6

3
Summer 2006 STAT 602 Homework 3 Solution

6. (a) The intervals (−∞, 3.5] and (3.5, 7.3] are disjoint. Thus,

F (7.3) = P (X ≤ 7.3) = P (−∞ < X ≤ 3.5 or 3.5 < X ≤ 7.3)


= P (−∞ < X ≤ 3.5) + P (3.5 < X ≤ 7.3) = F (3.5) + P (3.5 < X ≤ 7.3)

Therefore, P (3.5 < X ≤ 7.3) = F (7.3) − F (3.5) = 0.83 − 0.3 = 0.53

(b) 


 0.3, x=2



0.15, x=5






 0.38,

x = 7.3
f (x) =



 0.07, x=9


0.01, x = 100






0, elsewhere

4 2
7. P (A > B) = P (A = 4 and B = 3) = P (A = 4) · P (B = 3) = ·1=
6 3
4 2
P (B > C) = P (B = 3 and C = 2) = P (B = 3) · P (C = 2) = 1 · =
6 3

P (C > D) = P [(C = 2 and D = 1) or (C = 6 and D = 5) or (C = 6 and D = 1)]

= P (C = 2) · P (D = 1) + P (C = 6) · P (D = 5) + P (C = 6) · P (D = 1)

4 3 2 3 2 3
= · + · + ·
6 6 6 6 6 6
2
=
3
P (D > A) = P [(D = 1 and A = 0) or (D = 5 and A = 0) or (D = 5 and A = 4)]

= P (D = 1) · P (A = 0) + P (D = 5) · P (A = 0) + P (D = 5) · P (A = 4)

3 2 3 2 3 4
= · + · + ·
6 6 6 6 6 6
2
=
3
Z 4 4 4
(x + 2)2 (x + 2)3
Z
8. (a) E(X + 2) = (x + 2)f (x)dx = dx = =4
−2 −2 18 54 −2
Z 4 Z 4 4
(x + 2)3 (x + 2)4
(b) E (X + 2)2 = (x + 2)2 f (x)dx =
 
dx = = 18
−2 −2 18 72 −2
(c) E(X) = E(X) + 2 − 2 = E(X + 2) − 2 = 4 − 2 = 2

E (X + 2)2 = E[X 2 + 4X + 4] = E(X 2 ) + 4E(X) + 4


 

Thus 18 = E(X 2 ) + 8 + 4, E(X 2 ) = 6

4
Summer 2006 STAT 602 Homework 3 Solution
2
Therefore, V ar(X) = E(X 2 ) − (E(X)) = 6 − 4 = 2

9. Remark: It is not correct to attempt to solve this problem using integration by part, since that assumes
that the CDF is differentiable on [0, ∞). This assumption does not always hold.

The correct solution involves expressing the term 1 − F (x) as an integral, and then interchanging the
order of integration.
Z x
Let x ∈ [0, ∞), then F (x) = f (t)dt.
0
Z ∞
Since X is non-negative, f (t)dt = 1.
0
Z ∞ Z x Z ∞
So 1 − F (x) = f (t)dt − f (t)dt = f (t)dt.
0 0 x
Z ∞ Z ∞Z ∞
Hence, [1 − F (x)]dx = f (t)dtdx.
0 0 x

Since the region x < t < ∞, 0 < x < ∞ is equivalent to 0 < x < t, 0 < t < ∞, interchanging the order
of integration gives:
Z ∞ Z ∞ Z t
[1 − F (x)]dx = f (t)dxdt
0 0 0


Z ∞ Z ∞ Z t
[1 − F (x)]dx = f (t) 1dxdt
0 0 0
Z t
But 1dx = t. Therefore,
0
Z ∞ Z ∞
[1 − F (x)]dx = tf (t)dt = E(X).
0 0

Remark: A similar result holds when X is a non-negative random variable of the discrete type.

You might also like