A Simulation Study Of Ridge And Other Regression Estimators

Abstract

We consider a number of estimators of regression coefficients, all of generalized ridge, or ‘shrinkage1 type. Results of a simulation study indicate that with respect to two commonly used mean square error criteria, two ordinary ridge estimators, one proposed by Hoerl, Kennard and Baldwin, and the other introduced here, perform substantially better than both least squares and the other estimators discussed here. © 1976, Taylor & Francis Group, LLC. All rights reserved.