A test statistic to choose between Liu-type and least-squares estimator based on mean square error criteria

Abstract

In this study, the necessary and sufficient conditions for the Liu-type (LT) biased estimator are determined. A test for choosing between the LT estimator and least-squares estimator is obtained by using these necessary and sufficient conditions. Also, a simulation study is carried out to compare this estimator against the ridge estimator. Furthermore, a numerical example is given for defined test statistic.

Keywords:

• linear admissible estimators
• mean square error
• central-F approximations
• Liu-type estimator
• ridge estimator