Where X1: n × p’, and X2: n × (t−p ) and ϵ ∼ N(0, Iσ2). Suppose we estimate β1 and σ2 using the subset model
Under what conditions is2unbiased for σ2?
(c) Let X = ( X1X2) be of full column rank. Show that
(d) Using (c), show that the least squares estimators of the elements in β1, based on the subset model in (a), have smaller variance than the corresponding estimators of the full model.
Already registered? Login
Not Account? Sign up
Enter your email address to reset your password
Back to Login? Click here