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.
Original language | English (US) |
---|---|
Pages (from-to) | 179-211 |
Number of pages | 33 |
Journal | Journal of the Royal Statistical Society. Series B: Statistical Methodology |
Volume | 85 |
Issue number | 2 |
DOIs | |
State | Published - Apr 2023 |
Keywords
- causal inference
- censored data
- heterogeneous treatment effects
- machine learning
- random forest
- survival analysis
ASJC Scopus subject areas
- Statistics and Probability
- Statistics, Probability and Uncertainty