A0468
Title: Estimation of heterogeneous treatment effect using random forests for competing risks data
Authors: Youngjoo Cho - Konkuk University (Korea, South) [presenting]
Jaeseong Park - Korea University (Korea, South)
Abstract: The estimation of heterogeneous treatment effects for uncensored data has been studied extensively. However, efforts to develop methods of estimating heterogeneous treatment effect using machine learning has begun comparatively recently. A novel approach is proposed to estimating heterogeneous treatment effects with respect to cumulative incidence curves in the competing risks data using random forests. The proposed methods employ orthogonal estimating equations and augmented functions based on semiparametric efficiency theory. Simulation studies show the utility of this approach.
