EcoSta 2019: Start Registration
View Submission - EcoSta2019
Title: Group and individual variable selection in semiparametric transformation models Authors:  Jingjing Wu - University of Calgary (Canada) [presenting]
Wenyan Zhong - University of Calgary (Canada)
Xuewen Lu - University of Calgary (Canada)
Abstract: The bi-level variable selection is investigated in semiparametric transformation models for right-censored data. The class of transformation models under consideration includes the proportional hazards model and the proportional odds model as special cases and has the capability to accommodate external time-varying covariates. In the framework of regularized regression, we propose a computationally efficient estimation method that selects significant groups and variables simultaneously. Group bridge, adaptive group bridge and composite group bridge penalties which can integrate grouping structure of covariates were adopted for bi-level variable selection purpose. We illustrate the finite sample performance of the proposed methods via simulations and two real data examples.