EEEC10006: Probability Ming-Chun Lee

515000 Homework 7

Problem 1
Let X and Y be arbitrary real-valued random variables for which all expectations below
exist.

(a) Prove that  

Var(X) = E Var(X | Y ) + Var E[X | Y ] .
(4 pts)

(b) A certain amplifier’s output noise X (in watts) depends on a random operating
mode Y , which takes two values:

P(Y = 1) = p, P(Y = 2) = 1 − p.

Given Y = i, E[X | Y = i] = µi and Var(X | Y = i) = σi2 , for i = 1, 2.

(i) Use the law of total variance to compute Var(X) in terms of p, µ1 , µ2 , σ12 , σ22 .
(2 pts)
(ii) Explain in physical terms what each term E[Var(X | Y )] and Var(E[X | Y ])
represents in this circuit context. (4 pts)

Problem 2
Let MX (t) = E[etX ] denote the moment-generating function (MGF) of a random variable
X.

(a) You deposit $1 into a continuously-compounding account with interest rate t. The
random deposit time X follows an exponential distribution

X ∼ Exp(λ), fX (x) = λe−λx , x ≥ 0.

The terminal amount for a fixed deposit time τ can be derived by the discrete-
compounding approximation
 nτ
An (τ ) = 1 + nt , A(τ ) = lim An (τ ) = etτ .
n→∞

Consequently, for random deposit time X, the terminal amount is

A = etX .

Note: This problem makes the idealized assumption of truly continuous compound-
ing. In practice, interest is compounded at discrete intervals (e.g., daily, monthly,
or quarterly), not in a perfectly continuous manner

nτ
(i) Show the limit A(τ ) = lim 1 + nt = etτ . (2 pts)
n→∞

1
EEEC10006: Probability Ming-Chun Lee

(ii) Derive the MGF MX (t). (2 pts)

(iii) Explain what MX (t) represents physically in this compounding context. (2
pts)
(b) At temperature T , two independent resistors R1 and R2 generate thermal noise
voltages
V1 ∼ N (0, σ12 ), V2 ∼ N (0, σ22 ), σi2 = 4kT Ri B.
These noise sources are connected in series, yielding
Vseries = V1 + V2 .
(i) Calculate the MGFs of V1 , V2 , and Vseries . (2 pts)
Note for this subpart: For a normal random variable X ∼ N (µ, σ 2 ), you may
directly use
MX (t) = exp µt + 21 σ 2 t2 ,

and no further derivation is required.

(ii) Explain why MVseries (t) = MV1 (t) MV2 (t) corresponds to the physical fact that
voltages add in series and variances sum. (2 pts)

Problem 3
Calculate E[X 3 ] and E[X 4 ] for a standard normal random variable X. (10 pts)

Problem 4
A soccer team has three designated players who take turns striking penalty shots. The
ith player has probability of success pi , independent of the successes of the other players.
Let X be the number of successful penalty shots after each player has had one turn. Use
convolution to calculate the PMF of X. Confirm your answer by first calculating the
transform associated with X and then obtaining the PMF from the transform. (10 pts)

Problem 5
At a certain time, the number of people that enter an elevator is a Poisson random
variable with parameter λ. The weight of each person is independent of every other
person’s weight and is uniformly distributed between 100 and 200 lbs. Let Xi be the
fraction of 100 by which the ith person exceeds 100 lbs (for example, if the 7th person
weighs 175 lbs, then X7 = 0.75). Let
N
X
Y = Xi ,
i=1

where N ∼ Pois(λ) is the number of people entering the elevator.

(a) Find the transform associated with Y . (4 pts)
(b) Use the transform to compute E[Y ].(4 pts)
(c) Verify your answer to part (b) by using the law of iterated expectations. (2 pts)

2
EEEC10006: Probability Ming-Chun Lee

Problem 6
A pizza parlor serves k different types of pizza and is visited by a number H of customers
in each period, where H is a nonnegative integer random variable with a known associated
transform
MH (s) = E esH .
 

Each customer orders a single pizza, with all types of pizza being equally likely, indepen-
dent of the number of other customers and the types of pizza they order. Give a formula,
in terms of k and MH (·), for the expected number of different types of pizzas ordered.
(Write down your answer only in terms of k and MH (·).) (10 pts)

Problem 7
Consider a Gaussian (normal) random variable X with mean µx and variance σx2 . The
moment-generating function (MGF) of X is defined as

MX (t) = E etX .
 

(a) Derive the MGF of X. (2 pts)

(b) If the first moment and second moment of X are 4 and 20, respectively, find the
MGF MZ (t) of the random variable Z = X−62
, based on your result from (a). (2
pts)

(c) Based on the result obtained in (b), find the mean and variance of Z.(2 pts)

(d) Find the correlation coefficient between X and Z 2 . (2 pts)

(e) Find the probability Pr(Z ≥ −1). (2 pts)

Problem 8
Let X be a random variable with MGF MX (t), defined for −h < t < h. Determine which
of the following statements are true or false, and explain your answer.

(a) Pr(X ≥ 1) ≤ e−t MX (t), 0 < t < h. (3 pts)

(b) Pr(X ≤ 1) ≤ 1 − e−t MX (t), −h < t < 0. (3 pts)

(c) MX (t) + MX (6t), −h < t < h, is also the MGF of some random variable. (4 pts)

Problem 9
Given X1 , X2 , . . . , Xn , a sequence of i.i.d. random variables with expected value µx and
variance σx2 . Let
( ni=1 Xi ) − n µx
P
Zn = √ .
n σx
(a) Find E[Zn ] and Var(Zn ). (5 pts)

3
EEEC10006: Probability Ming-Chun Lee

(b) Show that as n → ∞, the CDF of Zn , denoted FZn (z), approaches the standard
normal CDF Φ(z). In other words,

lim FZn (z) = Φ(z).

n→∞

(5 pts)

Problem 10
Consider the calculation of the mean and variance of a sum

Y = X1 + · · · + XN ,

where N is itself a sum of integer-valued random variables, i.e.,

N = K1 + · · · + KL .

Here N , L, K1 , K2 , . . ., X1 , X2 , . . . are independent random variables. N , L, K1 , K2 , . . .

are integer-valued and nonnegative; K1 , K2 , . . . are i.i.d. with common mean and variance
denoted E[K] and Var(K); and X1 , X2 , . . . are i.i.d. with common mean and variance
denoted E[X] and Var(X).

(a) Derive formulas for E[N ] and Var(N ) in terms of E[L], Var(L), E[K] and Var(K).
(3 pts)

(b) Derive formulas for E[Y ] and Var(Y ) in terms of E[L], Var(L), E[K], Var(K), E[X]
and Var(X). ( 3 pts)

(c) A crate contains L cartons, where L is geometrically distributed with parameter q.

The ith carton contains Ki widgets, where Ki is Poisson-distributed with parameter
λ. The weight of each widget is exponentially distributed with parameter α. All
these random variables are independent. Find the expected value and variance of
the total weight of a crate. (4 pts)

Suggested Reading:
Chap.4:
Problem 16., Problem 21., Problem 26., Problem 28., Problem 29., Problem 33., Problem
35., Problem 36., Problem 39., Problem 42., Problem 45.