Let X_1, X_2, ..., X_n be iid observations from a N(u,σ^2) distribution with known variance.
Formulate the likelihood ratio test for testing Ho: u = 2 vs. Ho: u ≠2 (10 points).
Write down the equations needed to determine the sample size n and the critical value c such that the size of the test is 0.05 and the probability of type-II error when u = 2.5 is 0.1 (6 points) (note: only the equations are needed, no table checking is needed).
If we took 16 samples and observed X̄ = 2.5, and the population standard deviation is 0.5, compute the p-value based on the data for the test in a). Will you reject Ho at level α = 0.05 based on the p-value calculated? (4 points)
Hint: You can use
L(θ) = exp[-(n/2)ln(2πσ^2) - (1/2σ^2)∑(Xi - X̄)^2] exp[-(n/2)ln(2πσ^2) - (1/2σ^2)∑(Xi - u)^2] in part a) directly. In case you have difficulty deriving the rejection region in part a), please consider the rejection region |X̄ - u| > Zα/2(σ/√n) for part b).