Math 481 Spring 2021 Final exam
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
Math 481
Spring 2021
Final exam
1. Let x1 , . . . , xn be iid with population density
fX (①) = e- o(①)th(>)er(δ)ise.
Here δ, θ are unknown population parameters. You can use the fact that has Exp(1) distribution without proof in this problem.
(a) (1o points) Find the method of moments estimator for δ .
(b) (1o points) Find the method of moments estimator for θ .
(c) (1o points) Find the MLE estimator for δ .
(d) (1o points) Find the MLE estimator for θ .
2. (2o points) Let x1 , . . . , xn be iid N (1, σ2 ) where σ2 is an unknown parameter. consider the hypothesis test:
Ho : σ 2 = 4
H1 : σ 2 = 1.
construct the most powerful rejection region for this test with size a using Neyman-pearson lemma. You must do two things in this question: a) Determine the constant K of your region so that the size is a b) use Neyman-pearson to prove that your region is most powerful.
3. Let x1 , . . . , xn be iid uniform(o, β) where β is an unknown parameter. You can use the following fact without proof in this problem: if x, Y are independent uniform(o,1) then V = x + Y has density
fv (U) = {2(U) — U 1(o) 三(<) U(U) 2(1),
(a) (1o points) For n = 2 and a 三 1/2 ind k so that o < β < k(x1 + x2) is a 1 - a conidence interval for β .
(b) (1o points) For n = 2 and a 持 1/2 ind k so that o < β < k(x1 + x2) is a 1 - a conidence interval for β .
(c) (1o points) For n = 1oo, ind approximate value of k so that o < β < k Σ xi is a 1 - a conidence interval for β .
4. consider the Normal regression problem: Yi … N (a + β①i, σ2 ) where a, β, σ 2 are unknown. Recall the following:
S(S)从(从)从(Y)
where
S从从 = i(Σ)(①i - )2
S从Y = i(Σ)(①i - )(Yi - Y(-)).
(a) (5 points) show that S从Y = Σi(①i - )Yi.
(b) (5 points) show that Cov(Yi)Y(-)) = for all i.
na(a) ne-ay Hint(f Co):(v)apart(rian)ca(e) ob(s)and d(how th)m(at)ay(C)ou()) in(O.) this
2023-08-17