You are on page 1of 2

STAT 346 Probability for Engineers Spring 2018

George Mason University


Instructor: Martin Slawski Teaching Assistant: Xiaochen Zhu

Homework 7
due: Wednesday, April 25th, 2018 in class. Maximum possible points to earn: 14 [100%
= 12 points, rest counts as bonus]. All your answers need to be carefully justified to get
any credit.

Problem 1 (3 points)
Let (X, Y ) be two random variables with joint PDF
! !
x
fX,Y (x, y) = c · exp − , x, y ∈ R,

y
2

for some positive constant c > 0, where k·k2 denotes the Euclidean norm. Show that
Corr(X, Y ) = 0. Are X and Y independent? Explain your answer.

Hint. You can find E[XY ], E[X], and E[Y ] without integration, using the quadrant
symmetry of fX,Y .

Problem 2 (6 points)
Let X ∼ Bernoulli(p), Y ∼ Pois(λ) independent of X, and Z = XY .

a) Write down an expression for the PMF of Z.

b) Compute Corr(Y, Z).

c) Derive E[Z|X = x], x = 0, 1.

Problem 3 (5 points)
 
X
Consider the random vector  Y  with joint PDF
 

Z

c(x + y + 2z), 0 ≤ x ≤ 1, 0 ≤ y ≤ 1, 0 ≤ z ≤ 1,
fX,Y,Z (x, y, z) = 
0 else.
a)
Determine the value of the constant c.
b)
Compute P(X < 1/3 |Y < 1/2, Z < 1/4).

Hint. Use the definition of conditional probability and observe that

{Y < 1/2, Z < 1/4} = {0 < X < 1, 0 < Y < 1/2, 0 < Z < 1/4}.

You might also like