Modified GM6 Estimation method based on Improve Diagnostic
Robust Generalized Potential for quantile regression
Taha Alshaybawee
Mohammed Abdul Kareem
Al-Qadisiyah University – College of Administration and Economics
Corresponding Author: Mohammed Abdul Kareem
Abstract : In quantile regression, different methods have been suggested to mitigate the influence of high leverage
points. In this research paper, we will propose two new methods to robust quantile regression estimators by utilized
GM6- estimator and robust diagnostic method Improved Robust Generalized Potential which can be reduced masking
and swamping problems as well. Our proposed methods are GM6- for Quantile regression and GM6-for Quantile
regression based on Improved Diagnostic Generalized Potential, when the initial weights are calculated in the
algorithm for the last method based on Improved Diagnostic Generalized Potential method to detect high leverage
observation and reduce their impact in quantile regression. Simulations and real-world data were used to evaluate the
performance of our suggested methods to that of existing methods.
Keywords: leverage points, quantile regression, GM6-estimator, IDRGP, and GM6Q
