Local Learning for Covariate Selection in Nonparametric Causal Effect Estimation with Latent Variables

TL;DR

This paper introduces a local learning method for covariate selection in nonparametric causal effect estimation that effectively accounts for latent variables, improving efficiency and accuracy over global structure methods.

Contribution

The paper proposes a novel local learning approach that identifies valid covariate adjustment sets using independence tests, addressing limitations of existing global structure-based methods.

Findings

Effective covariate selection demonstrated on synthetic data.

Method outperforms global structure approaches in experiments.

Ensures soundness and completeness under standard assumptions.

Abstract

Estimating causal effects from nonexperimental data is a fundamental problem in many fields of science. A key component of this task is selecting an appropriate set of covariates for confounding adjustment to avoid bias. Most existing methods for covariate selection often assume the absence of latent variables and rely on learning the global network structure among variables. However, identifying the global structure can be unnecessary and inefficient, especially when our primary interest lies in estimating the effect of a treatment variable on an outcome variable. To address this limitation, we propose a novel local learning approach for covariate selection in nonparametric causal effect estimation, which accounts for the presence of latent variables. Our approach leverages testable independence and dependence relationships among observed variables to identify a valid adjustment set for a target causal relationship, ensuring both soundness and completeness under standard assumptions. We validate the effectiveness of our algorithm through extensive experiments on both synthetic and real-world data.