Suggest effective methods for solving the penalized least squares problem
min(y − Xb) T (y − Xb) + λ(b − b∗) T A(b − b∗)
for several values of λ at once.
Formulate the p = ∞ norm regression minb maxi |yi − xT i b| as a linear programming problem. Demonstrate the sensitivity of the resulting estimators to outliers.
Already registered? Login
Not Account? Sign up
Enter your email address to reset your password
Back to Login? Click here