Regularized Single Index Quantile Regression Model

摘要

This article proposes a new approach for variable selection in the single index quantile regression model. Compared to existing methods, the new approach produce sparse solutions for the index vector. Performance of the new method is enhanced by a fully adaptive penalty function. Finite sample performance is studied through a simulation study that compares the proposed method with existing work under several criteria. A data analysis is given which highlights the usefulness of the proposed methodology.