Adaptive LASSO Variable Selection Method For Current Status Data Under The Additive Hazards Model

Variable selection is commonly employed when the true underlying model has a sparse representation.Identifying significant covariates will enhance the prediction performance of the fitted model.To solve this problem,Zhang and Lu(2007)developed a variable selection method under the framework of the proportional hazards model when one observes right-censored data.In this paper,we consider the variable selection problem for the additive hazards model when one faces current status data.Motivated by Zhang and Lu(2007),we develop an adaptive LASSO method for this problem.The method is based on a penalized log partial likelihood with the adaptively weighted L₁penalty on regression coefficients.For variables with different degrees of importance,the penalties will be different and important variables tend to be retained in the selection process.Some theoretical properties,including consistency and oracle properties are established under some weak regularity conditions.Furthermore,we also provide an effective algorithm to solve this problem.An extensive simulation is performed to show that the method performs competitively.This method is also applied to a real data set from a tumorigenicity study.