STATS 100A Homework 4 Solution

STATS 100A Homework 4 Solution
Problem 1 Suppose we roll a die, let X be the number we get. Suppose p(x) = P (X = x) is such
that p(1) = .1, p(2) = .1, p(3) = .1, p(4) = .2, p(5) = .2, p(6) = .3.
(1) Calculate P (X > 4).
A: P (X > 4) = p(5) + p(6) = .2 + .3 = .5.
(2) Calculate P (X = 6|X > 4).
A: P (X = 6|X > 4) = P (X = 6)/P (X > 4) = .3/.5 = .6.
(3) Calculate E(X), Var(X), and SD(X).
A:
E(X) = 1 × 0.1 + 2 × 0.1 + 3 × 0.1 + 4 × 0.2 + 5 × 0.2 + 6 × 0.3 = 4.2

Var(X) = (1 − 4.2)2 × 0.1 + (2 − 4.2)2 × 0.1 + (3 − 4.2)2 × 0.1
+(4 − 4.2)2 × 0.2 + (5 − 4.2)2 × 0.2 + (6 − 4.2)2 × 0.3 = 2.76
p
SD(X) = Var(X) = 1.66
(4) Suppose the reward for x is h(x), and h(1) = −$20, h(2) = −$10, h(3) = $0, h(4) = $10,
h(5) = $20, h(6) = $100. Calculate E(h(X)), Var(h(X)) and SD(h(X)). What are the units of
E(h(X)) and Var(h(X))?
A:
E(h(X)) = −20 × 0.1 − 10 × 0.1 + 0 × 0.1 + 10 × 0.2 + 20 × 0.2 + 100 × 0.3 = 33$
Var(h(X)) = (−20 − 33)2 × 0.1 + (−10 − 33)2 × 0.1 + (0 − 33)2 × 0.1
+(10 − 33)2 × 0.2 + (20 − 33)2 × 0.2 + (100 − 33)2 × 0.3 = 2061$2
p
SD(h(X)) = Var(h(X)) = 45.4$
The units are attached in the above answers.

Problem 2 Suppose Z ∈ {0, 1}. P (Z = 1) = p, P (Z = 0) = 1 − p. Calculate E(Z), E(Z 2 ) and
Var(Z).
A: E(Z) = 1 × p + 0 × (1 − p) = p. Because Z 2 = Z, E(Z 2 ) = E(Z) = p. Var(Z) =
E(Z 2 ) − E(Z)2 = p − p2 = p(1 − p).
Problem 3 Suppose X is a discrete random variable following distribution p(x), where x takes
values in a discrete set.
(1) Prove E(aX) = aE(X).
A:
X X
E(aX) = axp(x) = a xp(x) = aE(X)
x x
(2) Prove E(X + b) = E(X) + b.

A:
X X X
E(X + b) = (x + b)p(x) = xp(x) + b p(x) = E(X) + b
x x x
(3) Prove Var(aX) = a2 Var(X).
1
A:
Var(aX) = E[aX − E(aX)]2 = E[aX − aE(X)]2

= E[a(X − E(X))]2 = a2 E[X − E(X)]2 = a2 Var(X)
(4) Prove Var(X + b) = Var(X).

A:
Var(X + b) = E[X + b − E(X + b)]2 = E[X + b − E(X) − b]2

= E[(X − E(X))]2 = E[X − E(X)]2 = Var(X)
(5) Prove Var(X) = E(X 2 ) − E(X)2 .

A:
Var(X) = E[X − E(X)]2 = E[X 2 − 2XE(X) + E(X)2 ] = E(X 2 ) − 2E(X)E(X) + E(X)2 = E(X 2 ) − E(X)2
(6) Let µ = E(X), and σ 2 = Var(X). Let Z = (X − µ)/σ. Calculate E(Z) and Var(Z).
A:
X −µ 1 1
E(Z) = E( ) = (E(X) − µ) = (µ − µ) = 0
σ σ σ
X −µ 1 1
Var(Z) = Var( ) = 2 Var(X − µ) = 2 Var(X) = 1
σ σ σ
Problem 4 Suppose we flip a fair coin 100 times independently. Let X be the number of heads.
Calculate E(X), Var(X), SD(X), E(X/100), Var(X/100), SD(X/100). Write down the formula
for computing P (X ∈ [40, 60]). Calculate the interval [E(X) − 2SD(X), E(X) + 2SD(X)].
A:
E(X) = np = 100 × (1/2) = 50

Var(X) = np(1 − p) = 100 × (1/2) × (1 − 1/2) = 25
p
SD(X) = Var(X) = 5
E(X/100) = E(X)/100 = 1/2
Var(X/100) = Var(X)/1002 = 2.5 × 10−3
SD(X/100) = SD(X)/100 = .05
60 60
X X 100
P (X ∈ [40, 60]) = P (X = k) = /2100
k
k=40 k=40
[40, 60]
Problem 5 Suppose within the population of voters, 20% of them support a candidate A. If
we randomly sample 100 people sequentially with replacement. Let X be the number of supporters
of A among these 100 people. Then what is the distribution of X? What are E(X), Var(X), and
SD(X)? What are E(X/100), Var(X/100), and SD(X/100)? Calculate the interval [E(X/100) −
2SD(X/100), E(X/100) + 2SD(X/100)].
2
A:
X ∼ Binomial(n = 100, p = .2)

E(X) = np = 100 × .2 = 20
Var(X) = np(1 − p) = 100 × .2 × (1 − .2) = 16
p
SD(X) = Var(X) = 4
E(X/100) = E(X)/100 = .2
Var(X/100) = Var(X)/1002 = 1.6 × 10−3
SD(X/100) = SD(X)/100 = .04
[.2 − 2 × .04, .2 + 2 × .04] = [.12, .28]
Problem 6 Suppose we randomly throw 10,000 points into the unit square [0, 1]2 . Let A be
the region x2 + y 2 ≤ 1. Let m be the number of points that fall into A. What is the distribution
of m? Let π̂ = 4m/10000 be our Monte Carlo estimate of π. What are E(π̂), Var(π̂) and SD(π̂)?
Calculate the interval [E(π̂) − 2SD(π̂), E(π̂) + 2SD(π̂)].
A:
m ∼ Binomial(n = 104 , p = π/4)

E(π̂) = E(4m/104 ) = (4/104 )E(m) = (4/104 ) × 104 × π/4 = π
Var(π̂) = Var(4m/104 ) = (4/104 )2 Var(m) = (4/104 )2 × 104 × π/4 × (1 − π/4) = π(4 − π)/104
p
SD(π̂) = π(4 − π)/102
p p
[π − 2 × π(4 − π)/102 , π + 2 × π(4 − π)/102 ]
Problem 7 Suppose a continuous random variable takes values within [0, 1], with a probability
density function f (x) = ax for x ∈ [0, 1], and f (x) = 0 otherwise. a is a positive constant.
(1) Calculate a.
R1
A: 0 axdx = 1. a = 2.
(2) Calculate F (x) = P (X ≤ x).
Rx
A: F (x) = 0 if x < 0. F (x) = 1 if x > 1. F (x) = 0 2xdx = x2 for x ∈ [0, 1].
(3) Calculate E(X) and Var(X).
R1 R1
A: E(X) = 0 xf (x)dx = 2/3. E(X 2 ) = 0 x2 f (x)dx = 1/2. Var(X) = E(X 2 ) − E(X)2 =
1/2 − 4/9 = 1/18.
(4) Please hand-draw a scatterplot of some points to illustrate the density.
3
4

STATS 100A Homework 4 Solution

Uploaded by

STATS 100A Homework 4 Solution

Uploaded by

STATS 100A Homework 4 Solution

E(X) = 1 × 0.1 + 2 × 0.1 + 3 × 0.1 + 4 × 0.2 + 5 × 0.2 + 6 × 0.3 = 4.2

The units are attached in the above answers.

(2) Prove E(X + b) = E(X) + b.

(3) Prove Var(aX) = a2 Var(X).

Var(aX) = E[aX − E(aX)]2 = E[aX − aE(X)]2

(4) Prove Var(X + b) = Var(X).

Var(X + b) = E[X + b − E(X + b)]2 = E[X + b − E(X) − b]2

(5) Prove Var(X) = E(X 2 ) − E(X)2 .

E(X) = np = 100 × (1/2) = 50

X ∼ Binomial(n = 100, p = .2)

m ∼ Binomial(n = 104 , p = π/4)

You might also like