To cheracterize the relationship between the response variable y and i covariates of interest, the following multiple linear regression model is used to fit the observed data: y=X3+ hyl Ble, B₁-, Bc. where y denotes the nx 1 response vector, 3 represents the px 1 parameter vector consisting of p=k+1 regression coefficients Bo, 31, 32,k, and e denotes the nx 1 vector of error terms. Assume that the model matrix X is an n x p full-column-rank matrix, and the entries of the first column of X are all equal to 1. In addition, assume that the error terms are independent and identically distributed normal random variables, that is, €12N (0,0³). Letz, denote the ith column of X. Suppose that a, sa for every i, where a represents a positive constant. Show that Var and the equality would be attained if X¹X aI, where 8, represents the ith entry of the least square estimator 3.
