Concordance and value information criteria for optimal treatment decision

Personalized medicine is a medical procedure that receives considerable scientific and commercial attention. The goal of personalized to assign the optimal treatment regime for each individual patient, according his/her personal prognostic information. When there are large number pretreatment variables, it crucial identify those important variables necessary decision making. In this paper, we study two information criteria: concordance value criteria, variable selection in We consider both fixed-$p$ high dimensional settings, show our criteria consistent model/tuning parameter selection. further apply four estimation approaches, including robust learning, concordance-assisted penalized A-learning sparse demonstrate empirical performance methods by simulations.