Open Access
March 2021 A Model Selection Approach for Variable Selection with Censored Data
María Eugenia Castellanos, Gonzalo García-Donato, Stefano Cabras
Author Affiliations +
Bayesian Anal. 16(1): 271-300 (March 2021). DOI: 10.1214/20-BA1207


We consider the variable selection problem when the response is subject to censoring. A main particularity of this context is that information content of sampled units varies depending on the censoring times. Our approach is based on model selection where all 2k possible models are entertained and we adopt an objective Bayesian perspective where the choice of prior distributions is a delicate issue given the well-known sensitivity of Bayes factors to these prior inputs. We show that borrowing priors from the ‘uncensored’ literature may lead to unsatisfactory results as this default procedure implicitly assumes a uniform contribution of all units independently on their censoring times. In this paper, we develop specific methodology based on a generalization of the g-priors, explicitly addressing the particularities of survival problems arguing that it behaves comparatively better than standard approaches on the basis of arguments specific to variable selection problems (like e.g. predictive matching) in the particular case of the accelerated failure time model with lognormal errors. We apply the methodology to a recent large epidemiological study about breast cancer survival rates in Castellón, a province of Spain.


The authors would like to thank the Epidemiological Unit and the cancer registry of Castellón of the Conselleria de Sanitat Universal i Salut Pública for sharing the breast cancer dataset.

The authors also thank the Editor, an Associate Editor and a referee for very valuable suggestions that have greatly improved the manuscript.


Download Citation

María Eugenia Castellanos. Gonzalo García-Donato. Stefano Cabras. "A Model Selection Approach for Variable Selection with Censored Data." Bayesian Anal. 16 (1) 271 - 300, March 2021.


Published: March 2021
First available in Project Euclid: 28 April 2020

MathSciNet: MR4194281
Digital Object Identifier: 10.1214/20-BA1207

Primary: 62C10 , 62C10
Secondary: 62F15

Keywords: Bayes factors , Bayesian model averaging , conventional priors , Model selection , objective priors , predictive matching

Vol.16 • No. 1 • March 2021
Back to Top