Causal Discovery via Statistical Power (CDSP)

2026-05-13

AI Summary

This work addresses the limited capacity of existing causal direction inference methods to effectively quantify uncertainty. The authors propose a bivariate causal discovery framework that integrates effect size estimation with statistical power analysis, systematically incorporating statistical power theory for the first time in this context. By formulating an asymmetry hypothesis based on effect sizes, the method enables causal direction identification with explicit uncertainty quantification. Combining effect size estimation, power analysis, and hypothesis testing, the approach demonstrates robustness to mild-to-moderate model misspecification in simulation studies. When evaluated on 100 real-world causal benchmark datasets, it reduces the false discovery rate by approximately 18% compared to current state-of-the-art methods.

Abstract

Causal discovery methods aim to infer causal direction from observational data. Functional causal discovery approaches use structural asymmetries to identify causal directionality but rely on strong modeling assumptions and provide limited tools for uncertainty quantification. We introduce Causal Discovery via Statistical Power (CDSP), a statistical inference framework that connects causal direction estimation with statistical power and enables uncertainty quantification. Considering the foundational setting of bivariate observational data, we show how quantities analogous to statistical power and effect size can be used in causal discovery to determine when data contain sufficient information to favor one direction over the other. We introduce the effect-size asymmetry assumption that characterizes when the probability of correctly detecting the causal direction (i.e., the power of causal discovery) exceeds that of incorrectly favoring the reverse direction. We show that the effect-size asymmetry assumption can be used for causal direction estimation with uncertainty quantification. Simulations show that CDSP direction estimation is robust to mild and moderate model misspecifications. Real data analyses on 100 cause-effect benchmark pairs further demonstrate that CDSP reduces false discovery rates by approximately 18% relative to a commonly used existing method.

Problem

• causal discovery
• causal direction
• uncertainty quantification
• observational data
• statistical power

Innovation

• causal discovery
• statistical power
• effect size asymmetry
• uncertainty quantification
• bivariate causal inference