Estimating heterogeneous treatment effects with right-censored data via causal survival forests
Yifan Cui, Michael R. Kosorok, Erik Sverdrup, Stefan Wager, Ruoqing Zhu
2023Journal of the Royal Statistical Society Series B (Statistical Methodology)92 citationsDOIOpen Access PDF
Abstract
Abstract Forest-based methods have recently gained in popularity for non-parametric treatment effect estimation. Building on this line of work, we introduce causal survival forests, which can be used to estimate heterogeneous treatment effects in survival and observational setting where outcomes may be right-censored. Our approach relies on orthogonal estimating equations to robustly adjust for both censoring and selection effects under unconfoundedness. In our experiments, we find our approach to perform well relative to a number of baselines.
Topics & Concepts
Censoring (clinical trials)Propensity score matchingObservational studyStatisticsEconometricsSurvival analysisSelection (genetic algorithm)Computer scienceEstimationParametric statisticsPopularityMathematicsMachine learningPsychologyEngineeringSystems engineeringSocial psychologyAdvanced Causal Inference TechniquesStatistical Methods and InferenceStatistical Methods and Bayesian Inference