Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: THEend8_
Stat 120B
Please read the following instructions carefully:
1. You must show your work/reasoning to receive full credit.
2. The exam is closed book and closed notes except for one 8.5”×11” sheet of notes, front
and back. You may use a calculator, but you may not use your cell phone/ipad/laptop
as a calculator.
3. During the ZOOM exam, please write your answers on separate papers (as you won’t
have access to a printer).
4. You may leave your answers unsimplified. For example, if the answer is
!
5
2
"
, you may
leave it in this form. However, if the answer involves an integral/derivative, you must
evaluate the integral/derivative.
5. There are five problems. Use your time wisely.
Honesty rules: No unauthorized notes, text, web searches, or other materials are allowed
during this examination. You are not allowed to communicate with anyone except your
TAs/Proctors and Prof. Qian during the exam. You are not allowed to disclose any part of
this exam to anyone. All work on this exam should be your own.
1
Question 1
Suppose X1 ∼ N(1, 5) and X2 ∼ N(−5, 9), and that X1 and X2 are independent. The
moment generating function (MGF) of N(µ, σ2) is M(t) = eµt+
σ2t2
2 . Use MGF to derive
the distribution of X1 + 2X2. (You should provide the name and the parameters of the
distribution as the final answer.)
2
Question 2
Consider X1, . . . , Xn that are independent and identically distributed, each with pdf
fX(x) = θx
θ−1, 0 < x < 1.
(a) Derive the likelihood function of θ given observed data xn = (x1, . . . , xn). Simplify as
far as possible.
(b) Find the maximum likelihood estimator (MLE) of θ. (You don’t need to check the
second-order derivative here.)
(c) Find the expectation of X, then find the method of moments estimator of θ.
(d) Is it possible for the true θ to be −1? Explain.
3
Question 3
LetX1, . . . , X80 be independent and identically distributed random variables following Uniform[−0.5, 0.5]
distribution. Let S = X1 + · · ·+X80.
(a) Let Z ∼ Uniform[0, 1]. We know that E(Z) = 0.5,Var(Z) = 1/12. It is easy to see that
(Z − 0.5) ∼ Uniform[−0.5, 0.5]. Use these results to derive E(X1) and Var(X1).
(b) Use the result in (a) to derive E(S) and Var(S).
(c) Use Chebyshev’s inequality to get an upper bound for P (|S| ≥ 4).
(d) What important distribution is the distribution of S very close to? Derive what the
parameters are, and state which theorem justifies your approximation.
(e) Use the approximate distribution for S in (d) to derive an approximation for P (|S| ≥ 4).
Express your answer using the CDF of N(0, 1), i.e., Φ(·).
4
Question 4:
Let X1, . . . , X9 be i.i.d. random variables following Poisson distribution Poisson(λ). The
PMF and MGF of Poisson(λ) are given by
P (X = x) =
λxe−λ
x!
, x = 0, 1, 2, . . . , MX(t) = exp(λ(e
t − 1)).
The mean and variance of this distribution is
E(X) = λ,Var(X) = λ.
(a) Let S =
#9
i=1 Xi. Find the distribution of S using MGF.
(b) Find the likelihood function of λ given observed data (x1, . . . , x9). Simplify as much as
possible.
(c) Find the maximum likelihood estimator (MLE) of λ given observed data (x1, . . . , x9).
(d) Consider an estimator λˆ = X¯n + 1 for λ. What are the bias, standard error, and MSE
of λˆ?
5
Question 5:
Suppose that the height of men has mean 68 inches and standard deviation 2.6 inches. We
draw 100 men at random (assume they are i.i.d.). Find (approximately) the probability that
the average height of men in our sample will be at least 68 inches.