On Some New Modifications of Ridge Estimators

TL;DR

This paper introduces new modifications of ridge estimators to improve multicollinearity handling, demonstrating superior performance through simulations and real data applications.

Contribution

It proposes novel ridge estimator modifications based on Lawless and Wang's approach, showing improved MSE performance over existing estimators.

Findings

New estimators outperform existing ones in most simulation scenarios

The proposed estimators show better MSE in real data applications

Monte Carlo experiments validate the improved performance

Abstract

Ridge estimator is an alternative to ordinary least square estimator when there is multicollinearity problem. There are many proposed estimators in literature. In this paper, we propose new estimators which are modifications of the estimator suggested by Lawless and Wang (1976). A Monte Carlo experiment has been conducted for the comparison of the performances of the estimators. Mean squared error (MSE) is used as a performance criterion. The benefits of new estimators are illustrated using two real datasets. According to both simulation results and applications, our new estimators have better performances in the sense of MSE in most of the situations.