CMStatistics 2017: Start Registration
View Submission - CMStatistics
Title: Envelope quantile regression Authors:  Shanshan Ding - University of Delaware (United States) [presenting]
Zhihua Su - University of Florida (United States)
Guangyu Zhu - University of British Columbia (Canada)
Lan Wang - University of Minnesota (United States)
Abstract: Quantile regression offers a valuable complement of classical mean regression for robust and comprehensive data analysis in a variety of applications. We propose a novel envelope quantile regression method (EQR) that adapts a nascent technique called enveloping to improve the efficiency of standard quantile regression. The new method aims to identify material and immaterial information in a quantile regression model and use only the material information for estimation without assuming the quantile regression coefficient vector is sparse. By excluding the immaterial part, the EQR method has the potential to substantially reduce the estimation variability with standard quantile regression. Unlike existing envelop model approaches which mainly rely on the likelihood framework, our proposed estimator is defined through a set of nonsmooth estimating equations. We facilitate the estimation via the generalized method of moments and derive the asymptotic normality of the proposed estimator by applying empirical process techniques. Furthermore, we establish that EQR is asymptotically more efficient than (or at least as asymptotically efficient as) the standard quantile regression estimators without imposing stringent conditions. Hence, the envelope model theory is advanced to general distribution-free settings. We demonstrate the effectiveness of the proposed method via Monte-Carlo simulations and a real data example.