Practically Effective Adjustment Variable Selection in Causal Inference

2025-01-09

Journal of Physics: Complexity

AI Summary

Causal effect estimation suffers from uncertainty in covariate selection—particularly under non-unique graph representations such as CPDAGs—and degraded accuracy in small-sample settings. Method: This paper proposes a practical, theoretically grounded criterion and algorithm for optimal adjustment set selection applicable to both DAGs and CPDAGs. It establishes, for the first time, a computability theorem for causal effects under CPDAGs, and designs an adjustment set selection criterion that jointly ensures identifiability and statistical efficiency, integrating the backdoor criterion, graph structure search, and rigorous theoretical justification. Results: Experiments on synthetic and real-world datasets demonstrate substantial improvements in causal effect estimation accuracy. The method exhibits strong robustness under small-sample conditions and when confounding structures are uncertain, providing a reliable, interpretable framework for covariate selection in data-limited causal inference.

Technology Category

Reasoning under Uncertainty: CausalityMachine Learning: Causal LearningConstraint Satisfaction and Optimization: Other Foundations of Constraint Satisfaction

Application Category

Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphsWeb Mining and Content Analysis: Robustness and generalizability of Web mining methodsSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

In the estimation of causal effects, one common method for removing the influence of confounders is to adjust the variables that satisfy the back-door criterion. However, it is not always possible to uniquely determine sets of such variables. Moreover, real-world data is almost always limited, which means it may be insufficient for statistical estimation. Therefore, we propose criteria for selecting variables from a list of candidate adjustment variables along with an algorithm to prevent accuracy degradation in causal effect estimation. We initially focus on directed acyclic graphs (DAGs) and then outlines specific steps for applying this method to completed partially directed acyclic graphs (CPDAGs). We also present and prove a theorem on causal effect computation possibility in CPDAGs. Finally, we demonstrate the practical utility of our method using both existing and artificial data.

Problem

• Selecting adjustment variables in causal inference
• Preventing accuracy degradation in causal effect estimation
• Applying method to directed and partially directed graphs

Innovation

• Algorithm prevents accuracy degradation
• Focuses on directed acyclic graphs
• Applies method to CPDAGs