Performance of Some Improved Estimators and Their Robust Versions Under Outliers: A Simulation Study

This paper introduces enhanced estimators designed to address the issue of multicollinearity in multiple linear regression models. In addition to multicollinearity, the presence of outliers also poses a significant challenge in regression analysis. To tackle these issues, this paper proposes several improved estimators, along with their robust counterparts, and compares their performance. The evaluation of these estimators is based on Monte Carlo simulations conducted under different levels of data contamination, ranging from clean datasets to moderate and heavy outlier presence. The mean squared error (MSE) is used as the performance criterion. The simulation results show that, in the absence of contamination, the improved estimators outperform most of their robust versions. However, when outliers are introduced at varying levels, the robust estimators consistently demonstrate superior performance compared to their conventional counterparts.