Treatment heterogeneity with right-censored outcomes using grf

TL;DR

This paper demonstrates how to estimate conditional average treatment effects for right-censored survival data using the causal_survival_forest function in the grf R package, illustrated with data from the National Job Training Partnership Act.

Contribution

It introduces a method for estimating treatment heterogeneity with censored outcomes using the grf package's causal_survival_forest function.

Findings

Effective estimation of CATEs with censored data demonstrated

Application to real-world job training data shown

Method improves understanding of treatment effects heterogeneity

Abstract

This article walks through how to estimate conditional average treatment effects (CATEs) with right-censored time-to-event outcomes using the function causal_survival_forest (Cui et al., 2023) in the R package grf (Athey et al., 2019, Tibshirani et al., 2024) using data from the National Job Training Partnership Act.