Practice: Compute the Kolmogorov-Smirnov Test Statistic 1 point possible (graded) Let X1, ..., Xn be iid...
KS Test Statistic 1/2 points (graded) In this problem, you will test the null and alternative hypotheses H0=the data set is distributed asUnif(0,1) H1=the data set is not distributed asUnif(0,1). What is the value of the Kolmogorov-Smirnov test statistic on the data set S? Enter TKS5/5–√, the KS statistic without the factor of n−−√, below. The problems on this page concern the data set S={0.28,0.2,0.01,0.80,0.1}. Let xi denote the i'th element of the data set S.
Consider X1,X2, , Xn be an iid random sample fron Unif(0.0). Let θ = (끄+1) Y where Y = max(X1, x. . . . , X.). It can be easily shown that the cdf of Y is h(y) = Prp.SH-()" 1. Prove that Y is a biased estimator of θ and write down the expression of the bias 2. Prove that θ is an unbiased estimator of θ. 3. Determine and write down the cdf of 0 4. Discuss why...
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.
Problem 4 Define f(x) as follows θ2 -1<=x<0 1-θ2 0<=x>1 0 otherwise Let X1, … Xn be iid random variables with density f for some unknown θ (0,1), Let a be the number of Xi which are negatives and b be the number of Xi which are positive. Total number of samples n = a+b. Find he Maximum likelihood estimator of θ? Is it asymptotically normal in this sample? Find the asymptotic variance Consider the following hypotheses: H0: X is...
Let X1, ..., Xn be IID observations from Uniform(0, θ). T(X) = max(X1, . . . Xn) is a sufficient statistic (additionally, T is the MLE for θ). Find a (1 − α)-level confidence interval for θ. [Note: The support of this distribution changes depending on the value of θ, so we cannot use Fisher’s approximation for the MLE because not all of the regularity assumptions hold.]
Let X1, X2 be iid, normal(µ, σ2 = 1). Show that the statistic T = X1 + X2 is sufficient for µ
Degrees of Freedom of a Known Test 2 points possible graded) Let us consider a statistical model with parameter ER". Let O be the parameter that generates the n lid samples X1,..., X, Let I ) be the Fisher information and assume that the MLE is asymptotically normal. Assume that I(C) is a diagonal matrix with positive entries 1/t1,...,1/td. We wish to perform a test for the hypotheses H : 8 - and H:8 + . Let the test statistic...
Let X1,…, Xn be a sample of iid random variable with pdf f (x; ?) = 1/(2x−?+1) on S = {?, ? + 1, ? + 2,…} with Θ = ℕ. Determine a) a sufficient statistic for ?. b) F(1)(x). c) f(1)(x). d) E[X(1)].
Let X1, X2,· · ·iid B(1, x), i.e,P(X1= 1) =x= 1−P(X1= 0), where x∈ [0,1]. Let Sn = X1+X2+· · ·+Xn. What can you say about the limiting behaviour of Sn/n from strong law large number
4.(120) Let X1,,,Xn be iid r(, 1) and g(u) given. Let 6n be the MLE of g(4) (1)(60) Find the asymptotic distribution of 6, (2)(60) Find the ARE of T Icc(X) w.r.t. on P(X1> c), c > 0 is i n i1 5.(80) Let X1, ,,Xn be iid with E(X1) = u and Var(X1) limiting distribution of nlog (1 +). o2. Find the where T n(X - 4)/s. - 1 - 4.(120) Let X1,,,Xn be iid r(, 1) and g(u)...