Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: THEend8_
STA 313
Some of these problems will be marked and posted on Crowdmark.
Problem 1 Let X be a r.v. having a normal distribution, N (a, σ2) with the parameters
EX = a and E(X − Ex)2 = σ2 > 0. Find P(X > a). Compute E (X −K)+, K > a.
Problem 2 Let ξ1, ξ2, ξn be i.i.d. random variables having uniform distribution in the in-
terval [0, 1]. Find the limit
lim
n→∞
(ξ1 · ξ2 · . . . · ξn)1/n .
Solve this problem using Monte Carlo method and analytically, using the Law of Large Num-
bers.
Problem 3 Let ξ1, ξ2, . . . ξn be i.i.d. random variables with mean 0 and finite variance,
0 < σ2 <∞. Prove that
n∑
k=1
ξk√
n∑
k=1
ξ2k
→ ξ in distribution as n→∞ ,
1
where ξ is a random variable having standard normal distribution.
Problem 4 Let X be a random variable, f and g are monotonically increasing, bounded
functions. Prove that f(X) and g(X) are positively correlated.
Hint: consider E
[
(f(X)− f(X ′)) · (g(X)− g(X ′))
]
, where X and X ′ are i.i.d.
Problem 5 (!) Let X and Y be r.v. such that EX2 < ∞ and EY 2 < ∞. Suppose
E(X |Y ) = Y and E(Y |X) = X. Prove that Y = X with probability 1.
Hint: compute E(X − Y )2 using conditioning on X and Y .
Problem 6 Let X be a normal N (0, 1) r.v.
i. Prove that
P(X > t) ≤ 1√
2pit
exp
(
−t
2
2
)
, t > 0.
ii. Compute E[Φ(aX + b)], where, as usual,
Φ(x) =
1√
2pi
∫ x
−∞
exp
(
−t
2
2
)
dt, a, b ∈ R.
Problem 7 Let X be a random variable having Poisson distribution with parameter λ > 0:
P (X = k) = e−λ
λk
k !
, k = 0, 1, 2, . . . .
Introduce a random variable Y such that Y = 0 if X is odd and Y = 1 if X is even. Find
E[X |Y ].
Problem 8 (Wald’s Identity) Let Xk be i.i.d. random variables such that E[X1] = m <
∞. Denote Sn =
∑n
k=1Xk. Let τ be an integer random variable independent of Xk. Prove
that
E
[ τ∑
k=1
Xk
]
= m · E[τ ].
Problem 9 Let Xn be i.i.d. with EXn = 0. Consider Sn =
n∑
k=1
Xk. Let τ ≥ 1 be a random
integer with finite mean, Eτ <∞. Prove that if σ2(Xn) <∞ then
σ2(Sτ ) = σ
2(X1) · Eτ.