2. (a) Fix po E (0, 1). Observing i.i.d. X1, ..., Xn Bernoulli(p) random variables where p E (0, 1), we test Ho : p = po against H1 : p / po. Instead of using the binomial distribution, use Gaussian approximation to derive the Wald test (z-test) at significance level a E (0, 1). (b) Suppose that we observe i.i.d. random variables X1, . .., Xn from a continuous distribution P with CDF F(t). Let M denote the median of P, that is, F(M) = P{X1