MSE Performance of a Heterogeneous Pre-Test Ridge Regression Estimator

Abstract

In this article, we consider a linear regression model and propose a heterogeneous pre-test ridge regression (HPRR) estimator which consists of the inequality constrained ridge regression (ICRR) estimator proposed by Ohtani (2008) and the ordinary least squares (OLS) estimator. We derive the exact formula for the mean squared error (MSE) of the HPRR estimator. Our numerical results show that the MSE performance of the estimator can be well controlled by introducing the pre-test.