Need to use differentiation to solve this problem.

5.10 Use the result of Problem 5.9 to derive a pair of linear equations for the least squares estimates of 1 and 2 for a causal AR(2) process (with mean zero). Compare your equations with those for the Yule-Walker estimates. (Assume that the mean is known to be zero in writing down the latter equations, so that the sample autocovariances are y (h) = _ _" " XithX, for h 2 0.) L ($, 02) = (2702) -1/2 ( det Gp) -1/2 result n x exp of 5.9 202 2 8, G,"X, + [ (X -QX-1 -... - Box-D)?|1. 1=p+1 where Xp = (X1, ..., Xp)' and Gp = 0 2 [p =0-2E(XX' )