Suppose Y1..... Y, are i.i.d. Poisson()). Consider the testing problem: Ho : x = > VS H1 : X # do. (a) Using the same notation as in the textbook, show that the (rescaled) log-likelihood ratio statistic is given by -2 log A(Y) = 2n (An - X) - Alog (do/A) where * = E; Vi. (b) Using Taylor expansion, prove the Wilk's phenomenon directly: -2log >(Y) -+d xi