Question: 2. Consider the linear regression models Full Model : Y = X+e; Y = e, Null Model: where XRxp is of full column rank.
2. Consider the linear regression models Full Model : Y = X+e; Y = e, Null Model: where XRxp is of full column rank. Assume that condition (N) holds. Let A = RPP be the identity matrix. Prove that (RSS null - RSS full)/(dfnu - df full) (A){A(XTX)A}A/p - 82 RSS full/df full where and are least squares estimators derived from the full model.
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
