Problem Set 6 Solution

HANU - Faculty of Information and Technology

MAT204: Probability & Statistics
April 10, 2015

Problem 1: [3, Exercise 3.75]

Let X and Y be random variables having joint density function

x + y 0 ≤ x ≤ 1, 0 ≤ y ≤ 1
f (x, y) =
0 otherwise.

Find (a) Var(X), (b) Var(Y ), (c) σX , (d) σY , (e) σXY , (f) ρ.
Solution:
Z +∞ Z +∞ Z 1 Z 1
E(X) = xf (x)dx = xf (x, y)dxdy = x(x + y)dxdy
−∞ −∞ x=0 y=0

1 1 1  1 1
x 3 x2 y
Z Z Z  Z  
2
 1 y
= (x + xy)dxdy = +  dy = + dy
x=0 y=0 y=0 3 2 
x=0 y=0 3 2
1
y y 2  1 1 7
= + = + =
3 4 y=0 3 4
 12
Z +∞ Z +∞ Z 1 Z 1
2 2 2
E(X ) = x f (x)dx = x f (x, y)dxdy = x2 (x + y)dxdy
−∞ −∞ x=0 y=0
1 1 1  1 Z 1 
x4 x3 y 
Z Z Z  
1 y
= (x3 + x2 y)dxdy = + dy = + dy
x=0 y=0 y=0 4 3 
x=0 y=0 4 3
1
y y 2  1 1 5
= + = + =
4 6 y=0 4 6
 12
Z +∞ Z +∞ Z 1 Z 1
E(Y ) = yf (y)dy = yf (x, y)dxdy = y(x + y)dxdy
−∞ −∞ x=0 y=0
1 1 1  1 1
y 3 xy 2
Z Z Z  Z  
2
 1 x
= (y + xy)dxdy = +  dx = + dx
x=0 y=0 x=0 3 2 
y=0 x=0 3 2
1
x x2  1 1 7
= + = + =
3 4 x=0 3 4
 12
Z +∞ Z +∞ Z 1 Z 1
2 2 2
E(Y ) = y f (y)dy = y f (x, y)dxdy = y 2 (x + y)dxdy
−∞ −∞ x=0 y=0

1
 1 1 1  1 1
y 4 xy 3
Z Z Z  Z  
3 2
 1 x
= (y + xy )dxdy = +  dx = + dx
x=0 y=0 x=0 4 3 
y=0 x=0 4 3
1
x x2  1 1 5
= + = + =
4 6 x=0 4 6
 12
Z ∞ Z ∞ Z 1 Z 1 Z 1 Z 1
E(XY ) = xyf (x, y)dxdy = xy(x+y)dxdy = (x2 y+xy 2 )dxdy
−∞ −∞ x=0 y=0 x=0 y=0
1  1 Z 1  1
x3 y x2 y 2  y y2 y2 y3
Z  
 =1+1=1

= + = + dy = +
y=0 3 2 
x=0 y=0 3 2 6 6 
y=0 6 6 3

a)
 2
2 5 2 7 11
V ar(X) = E(X ) − [E(X)] = − =
12 12 144

b)
 2
2 5 2 7 11
V ar(Y ) = E(Y ) − [E(Y )] = − =
12 12 144

c) r √
p 11 11
σX = V ar(X) = =
144 12

d) r √
p 11 11
σY = V ar(Y ) = =
144 12

e)   
1 7 7 1
σXY = Cov(X, Y ) = E(XY ) − E(X)E(Y ) = − =−
3 12 12 144

f)
Cov(X, Y ) σXY 1/144 1
p(X, Y ) = Cor(X, Y ) = = = −√ √ =−
σX σY σX σY 11/12 11/12 11

Problem 2: [3, Exercise 3.76]

Work Problem 1 if the joint density function is
 −(x+y)
e x ≥ 0, y ≥ 0
f (x, y) =
0 otherwise.

Solution: Z +∞ Z +∞ Z +∞
E(X) = xf (x, y)dxdy = xe−(x+y) dxdy = 1
−∞ x=0 y=0
Z +∞ Z +∞ Z +∞
E(X 2 ) = x2 f (x, y)dxdy = x2 e−(x+y) dxdy = 2
−∞ x=0 y=0

2
 Z +∞ Z +∞ Z +∞
E(Y ) = yf (x, y)dxdy = ye−(x+y) dxdy = 1
−∞ x=0 y=0
Z +∞ Z +∞ Z +∞
E(Y 2 ) = x2 f (x, y)dxdy = y 2 e−(x+y) dxdy = 2
−∞ x=0 y=0
Z +∞ Z +∞ Z +∞ Z +∞
E(XY ) = xyf (x, y)dxdy = xye−(x+y) dxdy = 1
−∞ −∞ x=0 y=0

a)
V ar(X) = E(X 2 ) − [E(X)]2 = 2 − 12 = 1

b)
V ar(Y ) = E(Y 2 ) − [E(Y )]2 = 2 − 12 = 1

c) p √
σX = V ar(X) = 1 = 1

d) p √
σY = V ar(Y ) = 1 = 1

e)
σXY = E(XY ) − E(X)E(Y ) = 1 − 1 = 0

f)
σXY 0
p(X, Y ) = = =0
σX σY 1∗1

Problem 3: [3, Exercise 3.79]

Find (a) the covariance, (b) The correlation coefficient of two random variables X and
Y if E(X) = 2, E(Y ) = 3, E(XY ) = 10, E(X 2 ) = 9, E(Y 2 ) = 16.
Solution:
a)
Cov(X, Y ) = σXY = E(XY ) − E(X)E(Y ) = 10 − 2 ∗ 3 = 4
b) √
V ar(X) = E(X 2 ) − [E(X)]2 = 9 − 22 = 5 ⇒ σX = 5
√
V ar(Y ) = E(Y 2 ) − [E(Y )]2 = 16 − 32 = 7 ⇒ σY = 7
σXY 4 4
Cor(X, Y ) = p(X, Y ) = =√ √ =√
σX σY 5 7 35

Problem 4: [3, Exercise 3.80]

1
The correlation coefficient of two random variables X and Y is − while their variances
4

3
are 3 and 5. Find the covariance.
Solution:
1
p(X, Y ) = −
4
2
√
V ar(X) = σX = 3 ⇒ σX = 3
√
V ar(Y ) = σY2 = 5 ⇒ σY = 5
√
σXY σXY 1 15
p(X, Y ) = = √ √ = − ⇒ σXY = −
σX σY 3 5 4 4

Problem 5: [3, Exercise 3.81]

Let X and Y have joint density function

x + y 0 ≤ x ≤ 1, 0 ≤ y ≤ 1
f (x, y) =
0, elsewhere.

Find the conditional expectation of (a) Y given X, (b) X given Y .

Solution:
a) 1
Z +∞ Z 1
y 2  1
f1 (x) = f (x, y)dy = (x + y)dy = xy + =x+
−∞ y=0 2 y=0 2
f (x, y) x+y 2(x + y)
f (y|x) = = =
f1 (x) x + 1/2 2x + 1
Z +∞ Z 1
2y(x + y)
E(Y |X) = yf (y|x)dy = dy
−∞ y=0 2x + 1
1
xy 2 2/3y 3  x + 2/3 3x + 2
= + = =
2x + 1 2x + 1 y=0 2x + 1
 6x + 3

b) 1
+∞ 1
x2
Z Z
 1
f2 (y) = f (x, y)dx = (x + y)dx = + xy  = y +
−∞ x=0 2 x=0 2
f (x, y) x+y 2(x + y)
f (x|y) = = =
f2 (y) y + 1/2 2y + 1
Z +∞ Z 1
2x(x + y)
E(X|Y ) = xf (x|y)dx = dx
−∞ x=0 2y + 1
1
x2 y 2/3x3  y + 2/3 3y + 2
= + = =
2y + 1 2y + 1 x=0 2y + 1
 6y + 3

Problem 6: [3, Exercise 3.82]

Work Problem 5 if
2e−(x+2y) x ≥ 0, y ≥ 0

f (x, y) =
0, elsewhere.

4
Solution:
a) Z +∞ Z 1
f1 (x) = f (x, y)dy = 2e−(x+2y) dy = e−x
−∞ y=0

f (x, y) 2e−(x+2y)
f (y|x) = = −x
= 2e−2y
f1 (x) e
Z +∞ Z ∞
1
E(Y |X) = yf (y|x)dy = 2ye−2y dy =
−∞ y=0 2

b) Z +∞ Z +∞
f2 (y) = f (x, y)dx = 2e−(x+2y) dx = 2e−2y
−∞ x=0

f (x, y) 2e−(x+2y)
f (x|y) = = = e−x
f2 (y) 2e−2y
Z +∞ Z +∞
E(X|Y ) = xf (x|y)dx = xe−x dx = 1
−∞ x=0

Problem 7: [3, Exercise 3.83]

Let X and Y have the joint probability function given in below table. Find the condi-
tional expectation of (a) Y given X, (b) X given Y .

Solution:
a)
P (X = 0, Y )
P (Y |X = 0) =
P (X = 0)
X 1 1 1 1
P (X = 0) = P (X = 0, Y ) = + + =
y
18 9 6 3

P (X = 0, Y = 0) 1/18 1
P (Y = 0|X = 0) = = =
P (X = 0) 1/3 6

5
 P (X = 0, Y = 1) 1/9 1
P (Y = 1|X = 0) = = =
P (X = 0) 1/3 3
P (X = 0, Y = 2) 1/6 1
P (Y = 2|X = 0) = = =
P (X = 0) 1/3 2
X 1 1 1 4
E(Y |X = 0) = y ∗ P (Y |X = 0) = 0 ∗ +1∗ +2∗ =
y
6 3 2 3

Similarly:
X 2 1 2
E(Y |X = 1) = y ∗ P (Y |X = 1) = 0 ∗ +1∗ +2∗ =1
y
5 5 5

X 3 3 1 5
E(Y |X = 2) = y ∗ P (Y |X = 2) = 0 ∗ +1∗ +2∗ =
y
7 7 7 7

X 0 1 2
E(Y |X) 4/3 1 5/7

b)
P (X, Y = 0)
P (X|Y = 0) =
P (Y = 0)
X 1 1 1 1
P (Y = 0) = P (X, Y = 0) = + + =
x
18 9 6 3
P (X = 0, Y = 0) 1/18 1
P (X = 0|Y = 0) = = =
P (Y = 0) 1/3 6
P (X = 1, Y = 0) 1/9 1
P (X = 1|Y = 0) = = =
P (Y = 0) 1/3 3
P (X = 2, Y = 0) 1/6 1
P (X = 2|Y = 0) = = =
P (Y = 0) 1/3 2
X 1 1 1 4
E(X|Y = 0) = x ∗ P (X|Y = 0) = 0 ∗ +1∗ +2∗ =
x
6 3 2 3

Similarly:
X 1 1 1 7
E(X|Y = 1) = x ∗ P (X|Y = 1) = 0 ∗ +1∗ +2∗ =
x
3 6 2 6

X 1 1 1 2
E(X|Y = 2) = x ∗ P (X|Y = 2) = 0 ∗ +1∗ +2∗ =
x
2 3 6 3

Y 0 1 2
E(X|Y ) 4/3 7/6 2/3

6
Problem 8: [3, Exercise 3.84]
Find the conditional variance of (a) Y given X, (b) X given Y for the distribution of
Problem 5.
Solution:
3y + 2
E(X|Y ) =
6y + 3
3x + 2
E(Y |X) =
6x + 3
2(x + y)
f (x|y) =
2y + 1
2(x + y)
f (y|x) =
2x + 1

a)
V ar(Y |X) = E(Y 2 |X) − [E(Y |X)]2
Z +∞ Z 1
2 (3x + 2)2 2 2(x + y) 9x2 + 12x + 4
= y f (y|x)dy − = y dy −
−∞ (6x + 3)2 y=0 2x + 1 (6x + 3)2
1
2/4y 4 + 2/3xy 3  9x2 + 12x + 4 1/2 + 2/3x 9x2 + 12x + 4
= − = −
2x + 1 
y=0 (6x + 3)2 2x + 1 9(2x + 1)2
6x2 + 6x + 1
=
18(2x + 1)2

b)
V ar(X|Y ) = E(X 2 |Y ) − [E(X|Y )]2
Z +∞ Z 1
2 (3y + 2)2 2 2(x + y) 9y 2 + 12y + 4
= x f (x|y)dx − = x dy −
−∞ (6y + 3)2 x=0 2y + 1 (6y + 3)2
1
2/4x4 + 2/3yx3  9y 2 + 12y + 4 1/2 + 2/3y 9y 2 + 12y + 4
= − = −
2y + 1 
x=0 (6y + 3)2 2y + 1 9(2y + 1)2
6y 2 + 6y + 1
=
18(2y + 1)2

Problem 9: [3, Exercise 3.85]

Work Problem 8 for the distribution of Problem 6.
Solution:
E(X|Y ) = 1
1
E(Y |X) =
2
f (x|y) = e−x
f (y|x) = 2e−2y

7
a)
V ar(Y |X) = E(Y 2 |X) − [E(Y |X)]2
Z +∞  2 Z +∞
1 1
= 2
y f (y|x)dy − = y 2 2e−2y dy −
−∞ 2 y=0 4
1 1 1
= − =
2 4 4

b)
V ar(X|Y ) = E(X 2 |Y ) − [E(X|Y )]2
Z +∞ Z +∞
= 2 2
x f (x|y)dx − 1 = x2 e−x dx − 1
−∞ x=0
=2−1=1

Problem 10: [3, Exercise 3.86]

Work Problem 8 for the distribution of the below table.

Solution:
a)
X 0 1 2
V ar(Y |X) 5/9 4/5 24/49

b)
Y 0 1 2
V ar(X|Y ) 5/9 29/36 5/9

Problem 11: [3, Exercise 3.105]

Let X be a random variable that can take on the values 2, 1, and 3 with respective
probabilities 1/3, 1/6, and 1/2. Find (a) the mean, (b) the variance, (c) the moment
generating function, (d) the characteristic function, (e) the third moment about the
mean.
Solution:

8
 x 2 1 3
f (x) 1/3 1/6 1/2

a)
X 1 1 1 7
E(X) = xf (x) = 2 ∗ +1∗ +3∗ =
x
3 6 2 3

b)
X 1 1 1
E(X 2 ) = + 1 2 ∗ + 32 ∗ = 6
x2 f (x) = 22 ∗
x
3 6 2
 2
2 2 7 5
V ar(X) = E(X ) − [E(X)] = 6 − =
3 9

c)
X 1 1 1 et + 2e2t + 3e3t
Mx (t) = E(etx ) = etx f (x) = e2t + et + e3t =
x
3 6 2 6

d)
eiω + 2e2iω + 3e3iω
ϕx (t) = E(eitx ) =
6

e)
µ3 = E[(X − µ)]3 = E(X 3 − 3X 2 µ + 3Xµ2 − µ3 )
= µ03 − 3µ02 µ + 3µ3 − µ3 = µ03 − 3µ02 µ + 2µ3
1 1 1 49
µ03 = E(X 3 ) = 23 ∗ + 13 ∗ + 33 ∗ =
3 6 2 3
 3
49 7 7 7
⇒ µ3 = −3∗6∗ +2 =−
3 3 3 27

Problem 12: [3, Exercise 3.106]

Work Problem 11 if X has density function

c(1 − x) 0 < x < 1
f (x) =
0 otherwise
where c is an appropriate constant.
Solution: Z +∞ 1
1
cx2 
Z
F (x) = f (x)dx = c(1 − x)dx = cx −
−∞ x=0 2 x=0
c c
=c− = =1⇒c=2
2 2

2(1 − x) 0 < x < 1
f (x) =
0 otherwise

a) 1
+∞ 1
2x3 
Z Z
22 1
E(X) = xf (x)dx = 2x(1 − x)dx = x − =1− =
−∞ x=0 3 x=0
 3 3

9
b)
+∞ 1
1
2x3 x4 
Z Z
2 2 2 1 12
E(X ) = x f (x)dx = 2x (1 − x)dx = − = − =
−∞ x=0 3 2 x=0 3 2
 6
 2
1 2 2 1 1
V ar(X) = E(X ) − [E(X)] = − =
6 3 18

c)
+∞ 1
2(et − 1 − t)
Z Z
tx tx
Mx (t) = E(e ) = e f (x)dx = etX 2(1 − x)dx =
−∞ x=0 t2

d)
−2(eiω − 1 − iω)
ϕx (t) = E(eitx ) =
ω2

e)
µ3 = E(X − µ)3 = µ03 − 3µ02 µ + 2µ3
Z 1 1
0 3 3 x4 2x5  1
µ3 = E(X ) = 2x (1 − x)dx = − =
x=0 2 5 x=0 10

 3
1 1 1 1 1
⇒ µ3 = −3∗ ∗ +2 =
10 6 3 3 135

Problem 13: [3, Exercise 3.109]

Let X and Y have joint density function

cxy 0 < x < 1, 0 < y < 1
f (x, y) =
0 otherwise
√
Find (a) E(X 2 + Y 2 ), (b) E( X 2 + Y 2 ).
Solution: Z +∞ Z +∞ Z 1 Z 1
F (x, y) = f (x, y)dxdy = cxydxdy
−∞ −∞ x=0 y=0
Z 1
1 Z 1
1
c 2  c c 2  c
= xy dy = ydy = y  = =1⇒c=4
y=0 2 x=0 y=0 2 4 Y =0 4

a)
Z 1 Z 1 Z 1 Z 1
2 2 2 2
E(X + Y ] = (x + y )f (x, y)dxdy = 4(x2 + y 2 )xydxdy
x=0 y=0 x=0 y=0

Z 1 Z 1 Z 1
1
3 3 4
 2 3
= (4x y + 4xy )dxdy = x y + 2x y  dy
x=0 y=0 y=0 x=0
1
1
y 2 1 4 
Z
1 1
3
= (y + 2y )dy = + y  = + =1
y=0 2 2 y=0 2 2

10
b)
p Z 1 Z 1 p Z 1 Z 1 p
2 2
E( (X + Y ) = 2 2
(x + y )f (x, y)dxdy = 4 (x2 + y 2 )xydxdy
x=0 y=0 x=0 y=0

Z 1
1 Z 1
2 2

2 3/2  4 h
3/2
3/2 i
=2 y(x + y )  dy = y 1 + y2 − y2 dy
y=0 3 x=0 y=0 3
1 √
2 2 2 5/2 4 5  4 √ 8 8(2 2 − 1)
= ∗ (1 + y ) − y  = ∗4 2− =
3 5 15 y=0 15 15 15

11
References
[1] Walpole, R. E., Myers, R. H., Myers, S. L. and Ye, K., Probability &
Statistics for Engineers & Scientists, 9th ed., MA, USA: Prentice-Hall, 2012.

[2] DeGroot, M. H. and Schervish, M. J., Probability and Statistics, 4th ed., MA,
USA: Pearson Education, Inc., 2012.

[3] Murray, R. S., John, J. S. and R, A. Srinivasan, Probability and Statistics,

3rd ed., USA: McGraw-Hill, 2009.

12