1

ISyE 6739 Practice Test #2 Solutions

Summer 2010
1. Short Answer Questions.
(a) Suppose X has p.d.f. f (x) = 5x4 , 0 x 1. Find E[2X 5].
ANSWER: 2E[X] 5 = 10/3.
(b) If X again has p.d.f. f (x) = 5x4 , 0 x 1, find Var(2X 5).
ANSWER: 4Var(X) = 0.0793.
(c) Suppose X can equal 1 or 2, each with probability 1/2. Find E[`n(X)].
ANSWER:

1
2

`n(2) = 0.347.

(d) TRUE or FALSE? E[X 2 ] (E[X])2 .

ANSWER: True.
(e) TRUE or FALSE? If Cov(X, Y ) = 0, then X and Y are independent.
ANSWER: False.
(f) Suppose that X and Y are independent Exponential() random variables.
Find Var(XY ) (yup the variance of the product).
ANSWER: E[X 2 Y 2 ] (E[XY ])2 = E[X 2 ]E[Y 2 ] (E[X])2 (E[Y ])2 = 3/4 .
(g) What does i.i.d. mean?
ANSWER: independent and identically distributed.
(h) Bonus Question: If you were Dutch, what would Q-65 mean to you?

2
ANSWER: They later became Golden Earring.
2. Suppose X and Y are two random variables such that E[X] = 3, Var(X) = 4,
E[Y ] = 2, Var(Y ) = 9, and Cov(X, Y ) = 2.
(a) Find the correlation between 2X and 3Y .
ANSWER:
Cov(X, Y )
= q
= 1/3.
Var(2X)Var(3Y )
Var(X)Var(Y )

Corr(2X, 3Y ) = q

Cov(2X, 3Y )

(b) Find Var(X Y ).

ANSWER: Var(X) + Var(Y ) 2Cov(X, Y ) = 17.
3. (10 points)
Suppose X has p.d.f. f (x) = 3x2 , 0 x 1. What is the p.d.f. of

Y = X?
ANSWER: The c.d.f. of Y is

G(y) = Pr(Y y) = Pr( X y)

2

= Pr(X y ) =

Z y2
0

3x2 dx

= y6.
Thus, the required p.d.f. is g(y) = 6y 5 , 0 y 1.
4. Suppose that f (x, y) = cxy if 0 < y < x < 1.
(a) Find c.
ANSWER: Use 1 =

R1Rx
0

cxy dy dx to get c = 8.

(b) Find fX (x).

ANSWER:

Rx
0

8xy dy = 4x3 , 0 < x < 1.

3
(c) Find Corr(X, Y ).
ANSWER: Using the usual methods, we have
4
,
5

E[X] =

Var(X) =

and
E[XY ] =
Thus,

2
,
75
Z 1Z x
0

Var(Y ) =

8x2 y 2 dy dx =

E[XY ] E[X]E[Y ]

8
,
15

E[Y ] =

Var(X)Var(Y )

11
,
225

4
.
9

= 0.492.

(d) Find f (y|x).

ANSWER: f (x, y)/fX (x) = 2y/x2 , 0 < y < x < 1.
(e) Use the above answer to find E[Y ].
ANSWER: First,
Z

E[Y |x] =

<

yf (y|x) dy =

Then

Z x
2y 2
0

x2

2x
.
3

dy =

E[Y ] = E[E[Y |X]] =

<

E[Y |x]fX (x) dx =

8
.
15

5. Suppose X is a discrete random variable whose only possible values are the nonnegative integers. Define the probability generating function of X by
gX (s) E[sX ] =

sk Pr(X = k).

k=0

(a) Prove E[X] =

d
ds

gX (s)|s=1 .

ANSWER:

d
d X
d
gX (s)|s=1 =
E[sX ]
= E
s
s=1
ds
ds
ds

= E Xs
s=1

X1

= E[X].
s=1

4
(b) If X Pois(), find gX (s). (You may need the fact that

k=0

y k /k! = ey .)

ANSWER:

gX (s) =

ke

k=0

k!

= e

X
(s)k
k=0

k!

= e(s1) .

(c) Suppose that X Pois(). Use the above two parts of the problem to find
E[X].
ANSWER:
E[X] =

gX (s)|s=1 = e(s1)
= .
s=1
ds

6. (Short normal distribution questions Just write your answer.)

(a) Find z0.025 .
ANSWER: 1.96.
(b) Find (1.85).
ANSWER: 0.9678.
(c) If X N (2, 4), find Pr{X < 4}.
ANSWER: Pr ( X2
<
2

42
)
2

= (1) = 0.8413.

(d) If X is standard normal, name the distribution of eX .

ANSWER: Lognormal.
(e) If X1 , X2 , . . . , X100 are i.i.d. Exp(1/3) random variables, what is the approxi
mate distribution of the sample mean, X?
Nor(3, 0.09).
ANSWER: By the Central Limit Theorem, X
(f) Suppose X1 , X2 , X3 are i.i.d. normal with mean 1 and variance 1. What is
Pr{2X1 X2 X3 > 0}?

5
ANSWER: 2X1 X2 X3 Nor(2 1 1, 4 + 1 + 1) Nor(0, 6), so an easy
symmetry argument says that the desired probability is 1/2.
7. Quickie Probability Questions Just Write Your Answer.
(a) If X is a continuous random variable with p.d.f. f (x) = (1/2)ex/2 for x 0,
find E[X].
ANSWER: 2.

(b) Suppose that X has m.g.f. MX (t) = pet + q. Whats E[X]?

ANSWER: MX0 (t) = pet |t=0 = p.

(c) TRUE or FALSE? The Law of the Unconscious Statistician states that, if g()
is a continuous function and X is a random variable, then E[g(X)] = g(E[X]).
ANSWER: False.

(d) TRUE or FALSE? If X and Y are independent continuous RVs, then

Pr(X Y ) =

Z Z y

f (x, y) dx dy.

ANSWER: True. 2
(e) TRUE or FALSE? E[E(X|Y )] = E[X].
ANSWER: True.

(f) If X Exp(3), what is its moment generating function?

ANSWER: 3/(3 t), for t < 3.

8. Suppose that Y is a random variable with p.d.f.

fY (y) = 4y 3 , 0 y 1.

6
Further suppose that the conditional p.d.f. of X given Y is
f (x|y) = 2x/y 2 , 0 x y 1.
Show your work on the following questions.
(a) Find f (x, y).
ANSWER: f (x, y) = f (x|y)fY (y) = 8xy, 0 x y 1.

(b) Find fX (x).

ANSWER: fX (x) =

R1
x

f (x, y) dy = 4x(1 x2 ), 0 x 1.

(c) Use Part 8b to find E[X].

ANSWER: E[X] =

R1
0

xfX (x) dx = 8/15.

(d) Find E[X|y].

ANSWER: E[X|y] =

Ry
0

xf (x|y) dx = 2y/3, 0 y 1.

(e) Use Part 8d to find E[X]. (This should match your answer in Part 8b.)
ANSWER: E[X] = E[E[X|Y ]] =
2

R1
0

E[X|y]fY (y) dy =

R1
0

(8/3)y 4 dy = 8/15.

(f) Find the correlation between X and Y .

ANSWER: After the usual algebra, we get E[X] = 8/15, E[X 2 ] = 1/3,
Var(X) = 11/225, E[Y ] = 4/5, E[Y 2 ] = 2/3, Var(Y ) = 2/75.
Further, E[XY ] =

R1R1
0

8x2 y 2 dy dx = 4/9, so that

E[XY ] E[X]E[Y ]
q

Var(X)Var(Y )

= 0.4924. 2

9. Given the following joint p.d.f.s (or other info), determine whether or not X and
Y must be independent. Just answer yes or no.

7
(a) g(x, y) = 2 e(x+y) , x > 0, y > 0.
ANSWER: Yes.

(b) f (x, y) = cxy, 0 < y < x < 1.

ANSWER: No.

(c) f (x, y) = c(1 + x + y)2 , x > 0, y > 0.

ANSWER: No.

(d) f (x|y) = fY (y) for all y.

ANSWER: No.

(e) Cov(X, Y ) = 0.
ANSWER: No.

10. Suppose that E(X) = 3, E(Y ) = 2, Var(X) = 5, Var(Y ) = 4, and Cov(X, Y ) = 2.

(a) Find E(2X + 3Y ).
ANSWER: 2E[X] + 3E[Y ] = 12.

(b) Find Var(2X + 3Y ).

ANSWER: 4Var(X) + 9Var(Y ) + 2Cov(2X, 3Y ) = 4Var(X) + 9Var(Y ) +
12Cov(X, Y ) = 32. 2
(c) Find the correlation between X and Y .
ANSWER: We have
Cov(X, Y )

= q

Var(X)Var(Y )

= 0.447. 2

8
11. Short Answer Questions.
(a) Suppose X has p.d.f. f (x) = 5x4 , 0 x 1. Find E[2X 5].
ANSWER: 2E[X] 5 = 10/3.
(b) If X again has p.d.f. f (x) = 5x4 , 0 x 1, find Var(2X 5).
ANSWER: 4Var(X) = 0.0793.
(c) Suppose X can equal 1 or 2, each with probability 1/2. Find E[`n(X)].
ANSWER:

1
2

`n(2) = 0.347.

(d) Suppose X has m.g.f. MX (t) = 0.3et + 0.7. Whats the distribution of X?
ANSWER: Bern(0.3).
(e) If X has m.g.f. 4/(4 t), for t < 4, find the m.g.f. of 2X 1.
ANSWER: 4et /(4 2t), t < 2.
(f) Suppose that X and Y are independent Exponential() random variables.
What is the m.g.f. of X + Y ?
ANSWER: (/( t))2 .
(g) Suppose X Exp(1/5). What is Pr(X > 7|X > 5).
ANSWER: By the memoryless property, Pr(X > 7|X > 5) = Pr(X > 2) =
e2/5 .
12. Dont forget to practice questions from your favorite distributions, e.g., from Homework 7 Bernoulli, binomial, geometric, negative binomial, Poisson, uniform, exponential, etc., etc.