Quantile Regression for Single-index Varying-coefficient Models with Missing Covariates at Random.

This paper studies the single-index varyingcoefficient quantile model with missing covariates at random. Firstly, some estimators of index parameters and their corresponding linkage function are given by using the inverse probability weighting method for missing data in two cases including parameter estimation and non-parametric estimation for the single-index varying-coefficient quantile regression model. In particular, the latter case focuses on the study of both known and unknown probability functions. Secondly, the established estimators are proved to be asymptotic normal under some suitable regularity conditions. Finally, the simulation studies are conducted to demonstrate the finite sample performance of the proposed method. [ABSTRACT FROM AUTHOR]