Detecting and Measuring Confounding Using Causal Mechanism Shifts

TL;DR

This paper introduces a comprehensive method for detecting and measuring both observed and unobserved confounding in causal inference without relying on strong parametric assumptions or causal sufficiency, leveraging recent advances in causal discovery.

Contribution

It relaxes traditional assumptions, proposes tailored methodologies for various confounding definitions, and introduces properties and measures for confounding that are supported by empirical validation.

Findings

Effective detection of confounding among variables

Ability to distinguish observed and unobserved confounding effects

Empirical validation supports theoretical properties

Abstract

Detecting and measuring confounding effects from data is a key challenge in causal inference. Existing methods frequently assume causal sufficiency, disregarding the presence of unobserved confounding variables. Causal sufficiency is both unrealistic and empirically untestable. Additionally, existing methods make strong parametric assumptions about the underlying causal generative process to guarantee the identifiability of confounding variables. Relaxing the causal sufficiency and parametric assumptions and leveraging recent advancements in causal discovery and confounding analysis with non-i.i.d. data, we propose a comprehensive approach for detecting and measuring confounding. We consider various definitions of confounding and introduce tailored methodologies to achieve three objectives: (i) detecting and measuring confounding among a set of variables, (ii) separating observed and unobserved confounding effects, and (iii) understanding the relative strengths of confounding bias between different sets of variables. We present useful properties of a confounding measure and present measures that satisfy those properties. Empirical results support the theoretical analysis.