Data adaptive covariate balancing for causal effect estimation for high dimensional data

2025-12-19

AI Summary

Estimating causal effects from high-dimensional observational data is hindered by the challenge of balancing confounders—existing weighting methods suffer from model misspecification, the curse of dimensionality, and failure to account for heterogeneous covariate importance. Method: We propose a nonparametric direct balancing framework that adaptively identifies true confounders via random forests, constructs a distance metric focused on shared covariates using leaf-node similarity, and enforces distributional balance in the L₂ sense via inverse-propensity-score convergence constraints. Contribution/Results: We theoretically prove that the learned weights converge consistently to the standardized inverse propensity scores. Empirically, our method achieves significantly lower bias and variance, and demonstrates superior robustness against model misspecification, outperforming state-of-the-art benchmark methods across diverse high-dimensional settings.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Intelligent Robots: State Estimation

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

A key challenge in estimating causal effects from observational data is handling confounding and is commonly achieved through weighting methods that balance distribution of covariates between treatment and control groups. Weighting approaches can be classified by whether weights are estimated using parametric or nonparametric methods, and by whether the model relies on modeling and inverting the propensity score or directly estimates weights to achieve distributional balance by minimizing a measure of dissimilarity between groups. Parametric methods, both for propensity score modeling and direct balancing, are prone to model misspecification. In addition, balancing approaches often suffer from the curse of dimensionality, as they assign equal importance to all covariates, thus potentially de-emphasizing true confounders. Several methods, such as the outcome adaptive lasso, attempt to mitigate this issue through variable selection, but are parametric and focus on propensity score estimation rather than direct balancing. In this paper, we propose a nonparametric direct balancing approach that uses random forests to adaptively emphasize confounders. Our method jointly models treatment and outcome using random forests, allowing the data to identify covariates that influence both processes. We construct a similarity measure, defined by the proportion of trees in which two observations fall into the same leaf node, yielding a distance between treatment and control distributions that is sensitive to relevant covariates and captures the structure of confounding. Under suitable assumptions, we show that the resulting weights converge to normalized inverse propensity scores in the L2 norm and provide consistent treatment effect estimates. We demonstrate the effectiveness of our approach through extensive simulations and an application to a real dataset.

Problem

• Develops nonparametric covariate balancing for causal estimation
• Addresses model misspecification and high-dimensional confounding challenges
• Uses random forests to adaptively emphasize true confounders

Innovation

• Uses random forests for nonparametric direct covariate balancing
• Jointly models treatment and outcome to identify confounders
• Constructs similarity measure from tree leaf node co-occurrence