Concordance-Assisted Learning for Estimating Optimal Individualized Treatment Regimes

• Activity Number: 431
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Caiyun Fan* and Wenbin Lu and Rui Song and Yong Zhou
• Companies: Shanghai University of International Business & Economics and North Carolina State University and North Carolina State University and Academy of Mathematics and Systems Science Chinese Academy of Sciences
• Keywords: Concordance ; Optimal treatment regime ; Propensity score ; Rank estimation ; Value function

Abstract

Zhang et al.(2012b) proposed a novel IPSW estimator for the mean potential outcome under a given treatment regime,i.e. the value function.The method is robust,however,the asymptotic distribution of the estimator of parameters is not normal and was not studied by Zhang et al.. Moreover,Zhang et al. proposed to use a genetic algorithm to maximize it,which may limit its practical applicability with relatively large-dimensional predictors.In this paper,we introduce a type of concordance function for prescribing treatment and propose a robust rank regression method for estimating the concordance function. We then find treatment regimes,up to a threshold,to maximize the concordance function,named prescriptive index.Finally,within the class of treatment regimes that maximize the concordance function,we find the optimal threshold to maximize the value function.We establish the consistency and limiting distribution of the proposed estimator for parameters.In addition,a doubly robust estimator of parameters in the prescriptive index is developed under a class of monotonic index models.The practical use and effectiveness of the proposed method are demonstrated by simulation and an AIDS data.