Orthogonal Random Forest for Causal Inference

Abstract

We propose the orthogonal random forest, an algorithm that combines Neyman-orthogonality to reduce sensitivity with respect to estimation error of nuisance parameters with generalized random forests (Athey et al., 2017)--a flexible non-parametric method for statistical estimation of conditional moment models using random forests. We provide a consistency rate and establish asymptotic normality for our estimator. We show that under mild assumptions on the consistency rate of the nuisance estimator, we can achieve the same error rate as an oracle with a priori knowledge of these nuisance parameters. We show that when the nuisance functions have a locally sparse parametrization, then a local ${\ell }_1$-penalized regression achieves the required rate. We apply our method to estimate heterogeneous treatment effects from observational data with discrete treatments or continuous treatments, and we show that, unlike prior work, our method provably allows to control for a high-dimensional set of variables under standard sparsity conditions. We also provide a comprehensive empirical evaluation of our algorithm on both synthetic and real data.

arXiv commentThis paper appeared in the Proceedings of the 36th International Conference on Machine Learning

Cited by in corpus (11)

• Double Machine Learning based Program Evaluation under Unconfoundedness
• Estimation and Inference with Trees and Forests in High Dimensions
• Estimating heterogeneous treatment effects with right-censored data via causal survival forests
• Localized Debiased Machine Learning: Efficient Inference on Quantile Treatment Effects and Beyond
• Kernel Conditional Moment Test via Maximum Moment Restriction
• Federated Estimation of Causal Effects from Observational Data
• Double/Debiased Machine Learning for Logistic Partially Linear Model
• Targeting predictors in random forest regression
• Split-Treatment Analysis to Rank Heterogeneous Causal Effects for Prospective Interventions
• Robust and Heterogenous Odds Ratio: Estimating Price Sensitivity for Unbought Items
• Adaptive Multi-Source Causal Inference