Variable Selection for Partially Linear Models with Missing Response at Random

摘要：

This paper presents a variable selection procedure by combining basis function approximations with penalized least-squares method for partially linear models with missing response at random. Based on local quadratic approximations, an iterative algorithm for finding the penalized estimators is proposed. Simulation results imply that the proposed variable selection method is workable.

关键词： partially linear model variable selection missing data penalized least-squares

作者: ZHAO Peixin

作者单位: Department of Mathematics, Hechi University, Guangxi, Yizhou, P.R.China, 546300

会议类型: 国际会议

会议名称: The 3rd International Institute of Statistics & Management Engineering Symposium(2010 国际统计与管理工程研讨会 IISMES)

会议地点: 威海

会议语种:英文

页码: 421-425

在线出版日期: 2010-07-24（万方平台首次上网日期，不代表论文的发表时间）