2012 Volume 39 Issue 1 Pages 9-23
This article presents flexible methods for modeling censored survival data using penalized smoothing splines when the covariate values change for the duration of the study. The Cox proportional hazards model has been widely used for the analysis of censored survival data. However, a number of theoretical problems with respect to the baseline survival function and the baseline cumulative hazard function remain unsolved. The basic concept considered in the present article is to use generalized additive models (GAM) with B-splines to estimate the survival function without the baseline hazard assumption. The proposed methods are discussed according to the way in which they deal with censored observations, competing risk, and time-dependent covariates. We evaluate the performance of the proposed method for predicting loan default with early payment as competing risk using data from a U.K. financial institution.