Variable selection for additive partially linear models with measurement error |
| |
Authors: | Zhangong Zhou Rong Jiang Weimin Qian |
| |
Institution: | 1.Department of Mathematics,Tongji University,Shanghai,People’s Republic of China;2.Department of Statistics,Jiaxing University,Zhejiang,People’s Republic of China |
| |
Abstract: | Variable selection for additive partially linear models with measurement error is considered. By the backfitting technique,
we first propose a variable selection procedure for the parametric components based on the smoothly clipped absolute deviation
(SCAD) penalization, and one-step spare estimates for parametric components are also presented. The resulting estimates perform
asymptotic normality as well as an oracle property. Then, two-stage backfitting estimators are also presented for the nonparametric
components by using the local linear method, and the structures of asymptotic biases and covariances of the proposed estimators
are the same as those in partially linear model with measurement error. The finite sample performance of the proposed procedures
is illustrated by simulation studies. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|