Back To Index Previous Article Next Article Full Text


Statistica Sinica 22 (2012), 531-554

doi:http://dx.doi.org/10.5705/ss.2010.134





PROFILED FORWARD REGRESSION FOR ULTRAHIGH

DIMENSIONAL VARIABLE SCREENING IN

SEMIPARAMETRIC PARTIALLY LINEAR MODELS


Hua Liang, Hansheng Wang and Chih-Ling Tsai


University of Rochester, Peking University and University of California, at Davis


Abstract: In partially linear model selection, we develop a profiled forward regression (PFR) algorithm for ultrahigh dimensional variable screening. The PFR algorithm effectively combines the ideas of nonparametric profiling and forward regression. This allows us to obtain a uniform bound for the absolute difference between the profiled predictors and their estimators. Based on this finding, we are able to show that the PFR algorithm uncovers all relevant variables within a few fairly short steps. Numerical studies are presented to illustrate the performance of the proposed method.



Key words and phrases: Forward regression, partially linear model, profiled forward regression, screening consistency, ultrahigh dimensional predictor.

Back To Index Previous Article Next Article Full Text