Implicitly Weighted Robust Estimation of Quantiles in Linear Regression

0508153 - ÚI 2020 CZ eng A - Abstract
Kalina, Jan - Vidnerová, Petra
Implicitly Weighted Robust Estimation of Quantiles in Linear Regression.
37th International Conference on Mathematical Methods in Economics 2019: Book of Abstracts. České Budějovice: 37th International Conference on Mathematical Methods in Economics 2019, 2019. s. 17-17.
[MME 2019: International Conference on Mathematical Methods in Economics /37./. 11.09.2019-13.09.2019, České Budějovice]
R&D Projects: GA ČR(CZ) GA19-05704S
Institutional support: RVO:67985807
Keywords : regression quantiles * robust regression * outliers * leverage points

Estimation of quantiles represents a very important task in econometric regression modeling and the standard regression quantiles machinery is well developed with numerous econometric applications. Although regression quantiles are commonly known as robust tools, they are vulnerable to the presence of leverage points in the data. We propose here a novel approach for the linear regression based on a specific version of the least weighted squares estimator, together with an additional estimator based only on observations between two different novel quantiles. Numerical computations reveal the new methods to perform comparably to standard regression quantiles, if the data are not contaminated by outliers. However, the new methods seem much more robust on a simulated dataset with severe leverage points. The new methods are also conceptually simple and comprehensible
Permanent Link: http://hdl.handle.net/11104/0299135