STA130B Problem 1 (1)
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
Problem 1 (14 points)
Let X1 , ..., Xn Laplace (θ) with pdf
f (xlθ) = e–9|z| (x e 皿 ; θ > 0)
NOTE: For this problem, denote the αth-quantile of the standard normal distribution as za , i.e. P (Z < za ) = α where Z ~ N (0, 1).
1. (7 pt) Compute the Fisher information I(θ).
2. (3 pt) Find the asymptotic distribution of θˆMLE .
3. (4 pt) Give an approximate 95% confidence interval for θ based on the asymptotic distri- bution of the MLE.
Problem 2 (20 points)
Let X1 , ..., Xn Laplace (θ) with pdf
f (xlθ) = e–9|z| (x e 皿 ; θ > 0)
NOTE: The pdf of the Exponential (λ) distribution is
f (xlλ) = λe–入z (x > 0 ; λ > 0)
The pdf of the Gamma (α, λ) distribution is
f (xlα, λ) = xa– 1 e –入z (x > 0; α > 0; λ > 0)
Note that the Exponential (λ) distribution is a special case of the Gamma (α, λ) distribution when α = 1.
1. (6 pt) Find a sufficient statistic for θ. Be sure to explicitly label the g and h functions when using the Factorization Theorem.
2. (10 pt) Let the prior distribution of θ be Exponential(λ) where the hyperparameter λ is known. Find and fully specify the posterior distribution of θ given an iid sample X1 , ..., Xn .
3. (4 pt) Write an integral expression for the posterior median. (You do not need to solve the integral).
Problem 3 (16 points)
Let X1 , ..., Xn Laplace (θ) with pdf
f (xlθ) = e–9|z| (x e 皿 ; θ > 0)
For θ 1 > θ0 , suppose we are testing the hypotheses
H0 : θ = θ0 vs. H1 : θ = θ 1
1. (2 pt) Explain why the Neyman-Pearson Lemma is an “optimality” result. In what way is a Neyman-Pearson based test the “best”?
2. (2 pt) Could we directly apply the Neyman-Pearson Lemma for the test described in this problem? Provide a short justification.
3. (6 pt) Show that a likelihood ratio test for H0 vs. H1 that rejects when Λ < C is equivalent to a test which rejects when lXi l < K for some constants C, K .
4. (6 pt) If X1 , ..., Xn Laplace (θ), then lXi l ~ Gamma(n, θ).
Let G0 ~ Gamma(n, θ0 ) and G1 ~ Gamma(n, θ1 ) with CDFs F0 and F1 respectively.
Write expressions for the significance level AND power of the test that rejects H0 when lXi l < K .
2022-02-23