Dimension reduction based linear surrogate variable approach for model free variable selection

doi:10.1016/j.jspi.2015.07.001

CSpace > 应用数学研究所

	Dimension reduction based linear surrogate variable approach for model free variable selection
	Dai, Pengjie 1; Ding, Xiaobo2 ; Wang, Qihua2,3
	2016-02-01
发表期刊	JOURNAL OF STATISTICAL PLANNING AND INFERENCE
ISSN	0378-3758
卷号	169 页码:13-26
摘要	Most of variable selection methods depend on the model assumptions, while sufficient dimension reduction is a nonparametric method to deal with high dimensional data. In this paper we aim at integrating sufficient dimension reduction into variable selection. A two stage procedure is proposed. First, we obtain dimension reduction directions and integrate them to construct a variable which is linearly dependent on predictors. Then by treating this constructed variable as a new response, we use the traditional variable selection methods such as adaptive LASSO to conduct variable selection. We call such a procedure as dimension reduction based linear surrogate variable (LSV) method. To illustrate that it has wide application, we also apply it to variable selection for the problem of missing responses. Extensive simulation studies show that it is more robust than the variable selection methods depending on model assumptions, and more efficient than the other model-free variable selection methods. Another advantage of the LSV is that it can be easily implemented. A real example is given to illustrate the proposed method. (C) 2015 Elsevier B.V. All rights reserved.
关键词	Adaptive LASSO Central subspace Linear surrogate variable Sufficient dimension reduction Variable selection
DOI	10.1016/j.jspi.2015.07.001
语种	英语
资助项目	Alzheimer's Disease Neuroimaging Initiative (ADNI) (National Institutes of Health)[U01 AG024904] ; National Institute on Aging ; National Institute of Biomedical Imaging and Bioengineering ; National Natural Science Foundation of China[11201457] ; National Natural Science Foundation of China[11171331] ; National Science Fund for Distinguished Young Scholars in China[10725106] ; National Science Fund for Creative Research Groups in China ; Natural Science Foundation ; Key Lab of Random Complex Structure and Data Science ; National Center for Mathematics and Interdisciplinary Sciences, CAS
WOS研究方向	Mathematics
WOS类目	Statistics & Probability
WOS记录号	WOS:000363827800002
出版者	ELSEVIER SCIENCE BV
引用统计
文献类型	期刊论文
条目标识符	http://ir.amss.ac.cn/handle/2S8OKBNM/21097
专题	应用数学研究所
通讯作者	Wang, Qihua
作者单位	1.Renmin Univ China, Sch Business, Beijing 100872, Peoples R China 2.Chinese Acad Sci, Acad Math & Syst Sci, Beijing 100190, Peoples R China 3.Shenzhen Univ, Inst Stat Sci, Shenzhen 518006, Peoples R China
推荐引用方式 GB/T 7714	Dai, Pengjie,Ding, Xiaobo,Wang, Qihua. Dimension reduction based linear surrogate variable approach for model free variable selection[J]. JOURNAL OF STATISTICAL PLANNING AND INFERENCE,2016,169:13-26.
APA	Dai, Pengjie,Ding, Xiaobo,&Wang, Qihua.(2016).Dimension reduction based linear surrogate variable approach for model free variable selection.JOURNAL OF STATISTICAL PLANNING AND INFERENCE,169,13-26.
MLA	Dai, Pengjie,et al."Dimension reduction based linear surrogate variable approach for model free variable selection".JOURNAL OF STATISTICAL PLANNING AND INFERENCE 169(2016):13-26.