Variable selection for semiparametric varying coefficient partially linear model based on modal regression with missing data |
| |
Authors: | Yafeng Xia Yarong Qu Nailing Sun |
| |
Institution: | 1. School of Sciences, Lanzhou University of Technology, Lanzhou, P. R. China;2. gsxyf01@163.com;4. School of Foreign Languages, Lanzhou University of Technology, Lanzhou, P. R. China |
| |
Abstract: | AbstractIn this article, we focus on the variable selection for semiparametric varying coefficient partially linear model with response missing at random. Variable selection is proposed based on modal regression, where the non parametric functions are approximated by B-spline basis. The proposed procedure uses SCAD penalty to realize variable selection of parametric and nonparametric components simultaneously. Furthermore, we establish the consistency, the sparse property and asymptotic normality of the resulting estimators. The penalty estimation parameters value of the proposed method is calculated by EM algorithm. Simulation studies are carried out to assess the finite sample performance of the proposed variable selection procedure. |
| |
Keywords: | Semiparametric varying coefficient partially linear model missing data modal regression variable selection |
|
|