Variable selection in Cox regression models with varying coefficients

Toshio Honda (Graduate School of Economics, Hitotsubashi University)
Monday, June 25, 2012 - 2:00pm
Spandauer Strasse 1, Room 23

I deal with two kinds of Cox regression models with varying coefficients. The coefficients vary with time in one model. In the other model, there is an exposure covariate and the coefficients vary with the exposure variable. In both models, we have p-dimensional covariates and p increases moderately. However, only a small part of the covariates may be significant. We carry out estimation and variable selection by using the group SCAD-type estimator and the adaptive group Lasso estimator and examine the theoretical properties of the estimators, especially the sparsity and the oracle property.