Comparison of Estimation Methods for Regression Model Parameters in the Case of the Problem of Linear Multiplicity and Abnormal Values

Authors

• Nazar Mustafa Alsarraf, Department of Statistics, College of Administration and Economics, University of Baghdad, Iraq
• Ghufran I. Kaml, Department of Statistics, College of Administration and Economics, University of Baghdad, Iraq

Abstract

A simulation study is used to examine the robustness of some estimators on a multiple linear regression model with problems of multicollinearity and non-normal errors, the Ordinary least Squares (LS) ,Ridge Regression, Ridge Least Absolute Value (RLAV), Weighted Ridge (WRID), MM and a robust ridge regression estimator MM estimator, which denoted as RMM this is the modification of the Ridge regression by incorporating robust MM estimator . finialy, we show that RMM is the best among the other estimators

DOI

10.33095/jeas.v15i55.1260

Subject Area

Statistical

First Page

153

Last Page

166

Recommended Citation

Alsarraf, N. M., & Kaml, G. I. (2009). Comparison of Estimation Methods for Regression Model Parameters in the Case of the Problem of Linear Multiplicity and Abnormal Values. Journal of Economics and Administrative Sciences, 15(55), 153-166. https://doi.org/10.33095/jeas.v15i55.1260