Overview
In this lecture we look at ridge regression from a Bayesian perspective and discuss choice of priors and inference via MCMC
Readings
Readings: Christensen Chapter 15 and Hoff Chapter 9
In this lecture we look at ridge regression can be formulated as a Bayesian estimator and discuss prior distributions on the ridge parameter. As estimators with smaller MSE can be obtained by allowing a different shrinkage parameter for each coordinate we relax the assumption of a common ridge parameter and consider generalized ridge estimators and implications for prior choice.