Variable selection for interval-censored failure time data

Abstract

[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT AUTHOR'S REQUEST.] Variable selection is a commonly asked question and various traditional variable selection methods have been developed, including forward, backward and stepwise selection, as well as best subset selection. Among these conventional selection techniques, the best subset selection is recommended by most researchers. However, this method requires fitting all sub-models, which can be very time-consuming when the number of covariates p is large.