TY - JOUR
T1 - Feature selection for varying coefficient models with ultrahigh-dimensional covariates
AU - Liu, Jingyuan
AU - Li, Runze
AU - Wu, Rongling
N1 - Funding Information:
Jingyuan Liu is Assistant Professor, Department of Statistics, School of Economics, Wang Yanan Institute for Studies in Economics and Fujian Key Laboratory of Statistical Science, Xiamen University, China (E-mail: [email protected]). Runze Li is Distinguished Professor, Department of Statistics and The Methodology Center, The Pennsylvania State University, University Park, PA 16802-2111 (E-mail: [email protected]). Rongling Wu is Professor, Department of Public Health Sciences, Penn State Hershey College of Medicine, Hershey, PA 17033 (E-mail: [email protected]). The research of Runze Li was supported by National Institute on Drug Abuse (NIDA) grant P50-DA10075, National Cancer Institute (NCI) grant R01 CA168676, and National Natural Science Foundation of China grant 11028103. The research of Rongling Wu was supported by an NSF grant IOS-0923975 and an NIH grant UL1RR0330184. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NSF, NIH, NIDA, and NCI.
PY - 2014
Y1 - 2014
N2 - This article is concerned with feature screening and variable selection for varying coefficient models with ultrahigh-dimensional covariates. We propose a new feature screening procedure for these models based on conditional correlation coefficient. We systematically study the theoretical properties of the proposed procedure, and establish their sure screening property and the ranking consistency. To enhance the finite sample performance of the proposed procedure, we further develop an iterative feature screening procedure. Monte Carlo simulation studies were conducted to examine the performance of the proposed procedures. In practice, we advocate a two-stage approach for varying coefficient models. The two-stage approach consists of (a) reducing the ultrahigh dimensionality by using the proposed procedure and (b) applying regularization methods for dimension-reduced varying coefficient models to make statistical inferences on the coefficient functions. We illustrate the proposed two-stage approach by a real data example. Supplementary materials for this article are available online.
AB - This article is concerned with feature screening and variable selection for varying coefficient models with ultrahigh-dimensional covariates. We propose a new feature screening procedure for these models based on conditional correlation coefficient. We systematically study the theoretical properties of the proposed procedure, and establish their sure screening property and the ranking consistency. To enhance the finite sample performance of the proposed procedure, we further develop an iterative feature screening procedure. Monte Carlo simulation studies were conducted to examine the performance of the proposed procedures. In practice, we advocate a two-stage approach for varying coefficient models. The two-stage approach consists of (a) reducing the ultrahigh dimensionality by using the proposed procedure and (b) applying regularization methods for dimension-reduced varying coefficient models to make statistical inferences on the coefficient functions. We illustrate the proposed two-stage approach by a real data example. Supplementary materials for this article are available online.
UR - http://www.scopus.com/inward/record.url?scp=84901807757&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84901807757&partnerID=8YFLogxK
U2 - 10.1080/01621459.2013.850086
DO - 10.1080/01621459.2013.850086
M3 - Article
C2 - 24678135
AN - SCOPUS:84901807757
SN - 0162-1459
VL - 109
SP - 266
EP - 274
JO - Journal of the American Statistical Association
JF - Journal of the American Statistical Association
IS - 505
ER -