be sure to 1. [Derivation/Conceptual] Consider the multiple linear regression model with the usual setup, where Y is the n x 1 column vector of responses, X is the n x (p + 1) matrix for the predictors (with intercept), and ?' ~ MV N(0, o2 Inxn). Recall that we have the estimator # = (X?X)-1XTY, and we defined the "hat" matrix H = X(XTX)-1XT. Viewed as random vectors, the fitted values are / = XS and the residuals are e = Y - ji. (a) State the distributions of / and