Path-specific Effects Based on Information Accounts of Causality

TL;DR

This paper introduces a novel path intervention method based on information accounts of causality, providing a clearer interpretation and broader applicability for path-specific effects in mediation analysis.

Contribution

It proposes a new path intervention framework inspired by information theory, with explicit manipulation description and no need for non-existence of recanting witness.

Findings

Develops intervention diagrams and $$-formula for the new approach

Explicitly describes manipulation in structural causal models

Does not require non-existence of recanting witness for identification

Abstract

Path-specific effects in mediation analysis provide a useful tool for fairness analysis, which is mostly based on nested counterfactuals. However, the dictum ``no causation without manipulation'' implies that path-specific effects might be induced by certain interventions. This paper proposes a new path intervention inspired by information accounts of causality, and develops the corresponding intervention diagrams and $\pi$-formula. Compared with the interventionist approach of Robins et al.(2020) based on nested counterfactuals, our proposed path intervention method explicitly describes the manipulation in structural causal model with a simple information transferring interpretation, and does not require the non-existence of recanting witness to identify path-specific effects. Hence, it could serve useful communications and theoretical focus for mediation analysis.