3. Suppose that we have two independent random samples: X1, Xn are exponential(), and Y1,... ....
4. (30 points) Suppose that we have two independent random samples: X1, X2, ..,,Xn are exponential (9) and Y.Y2, ,,Yn are exponential(A) (aside: be happy l didn't make it(!) a. Find the likelihood ratio test of Ho: θ 1 versus H1 : θ . b. Show that the test in part a. can be based on the statistic ΣΑΜ c. Find the distribution of T when Ho is true.
Please justify each step! 4. (30 points) Suppose that we have two independent random samples: X1, X2, ...,, Xn are exponential(8) and Y. Y, , , Yn are exponential(A) (aside: be happy I didn't make it 〈!) a. Find the likelihood ratio test of Ho: θ μ versus H1:0 # . b. Show that the test in part a. can be based on the statistic c. Find the distribution of T when Ho is true. 4. (30 points) Suppose that...
Let X1, . . . , Xn ∼ iid Exp(λ) and Y1, . . . , Ym ∼ iid Exp(τ ) be independent random samples. (a) Find the restricted MLEs under the null hypothesis H0 : λ = τ . (b) Write out a formula for the LRT statistic, and describe how you could perform this test asymptotically.
Given two independent random samples X1, ..., Xn and Y1, ..., Ym with normal dis- tributions N(Hz, o?) and N(Hy, oz), determine a generalized likelihood ratio test for Ho : Mix - My = 0 versus H : plz – My 70 at a given significance level a (01, 0y unknown but equal).
5. We have two independent samples of n observations X1,X2-…Xn and Yi,½, . …Ý, We want to test the hypothesis H0 : μ®-ty versus the alternative H1 : μζ μυ. (a) First, assume that the null hypothesis H0 is true and find the MLE for μ- - y. (b) Then plug this estimate into the log likelihood along with the MLBs μτ-x and μ to calculate the LRT statistic. (c) Is this likelihood ratio test equivalent to the test that...
Suppose that X1, X2, ..., Xn are independent random variables (not iid) with densities ÍXi(z10,) -.2 e _ θ:/z1(z > 0), where θί 〉 0, for i = 1, 2, , n. (a) Derive the form of the likelihood ratio test (LRT) statistic for testing versuS H1: not Ho. You do not have to find the distribution of the likelihood ratio test (LRT) statistic under Ho- Just find the form of the statistic. (b) From your result in part (a),...
5. We have two independent samples of n observations X1, X2, .. . , Xn and Yı, Y2,.. . , Yn We want to test the hvpothesis H 0 : μΧ-My versus the alternative H1 : μΧ * My (a) First, assume that the null hypothesis Ho is true and find the MLE for μ-Ac-My (b) Then plug this estimate into the log likelihood along with the MLE's μχ-x and My-- to calculate the LRT statistic (c) Is this likelihood...
Suppose that X1, X2,.... Xn and Y1, Y2,.... Yn are independent random samples from populations with the same mean μ and variances σ., and σ2, respectively. That is, x, ~N(μ, σ ) y, ~ N(μ, σ ) 2X + 3Y Show that is a consistent estimator of μ.
2. Let X1, , Xn be iid exponential(9) random variables. Derive the LRT of Ho : ? = ?? versus Ha : ????. Determine an approximate critical value for a size-a test using the large sample approximation.
Let X1, ..., Xn and Y1, ..., Ym be two independent samples from a Poisson dis- tribution with parameter 1. Let a, b be two positive numbers. Consider the following estimator for 1: i ,Y1 +...+Ym = a- X1 +...+Xn n т (a) What condition is needed on a and b so that û is unbiased? (b) What is the MSE of i?