Consider the MA(1) model Xt = Zt + Zt1, R, where Zt are i.i.d. random variables with mean 0 and variance 2 Z. Our goal is to find the best linear predictor PnXn+1of Xn+1 based on X1, . . . , Xn. (a) Let n = 1. Use the formula nan = (n; 1) to conclude that P1X2 = X(1) X(0)X1 = 1 + 2 X1. (b) Let n = 2. Use the formula nan = (n; 1) to obtain coefficients a1, a2 in P2X3 = a1X2 + a2X1.