Estimation and Inference for Causal Explainability

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Weihan, Gao, Zijun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912946991398912
author Zhang, Weihan
Gao, Zijun
author_facet Zhang, Weihan
Gao, Zijun
contents Understanding how much each variable contributes to an outcome is a central question across disciplines. A causal view of explainability is favorable for its ability in uncovering underlying mechanisms and generalizing to new contexts. Based on a family of causal explainability quantities, we develop methods for their estimation and inference. In particular, we construct a one-step correction estimator using semi-parametric efficiency theory, which explicitly leverages the independence structure of variables to reduce the asymptotic variance. For a null hypothesis on the boundary, i.e., zero explainability, we show its equivalence to Fisher's sharp null, which motivates a randomization-based inference procedure. Finally, we illustrate the empirical efficacy of our approach through simulations as well as an immigration experiment dataset, where we investigate how features and their interactions shape public opinion toward admitting immigrants.
format Preprint
id arxiv_https___arxiv_org_abs_2512_20219
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Estimation and Inference for Causal Explainability
Zhang, Weihan
Gao, Zijun
Methodology
Applications
Understanding how much each variable contributes to an outcome is a central question across disciplines. A causal view of explainability is favorable for its ability in uncovering underlying mechanisms and generalizing to new contexts. Based on a family of causal explainability quantities, we develop methods for their estimation and inference. In particular, we construct a one-step correction estimator using semi-parametric efficiency theory, which explicitly leverages the independence structure of variables to reduce the asymptotic variance. For a null hypothesis on the boundary, i.e., zero explainability, we show its equivalence to Fisher's sharp null, which motivates a randomization-based inference procedure. Finally, we illustrate the empirical efficacy of our approach through simulations as well as an immigration experiment dataset, where we investigate how features and their interactions shape public opinion toward admitting immigrants.
title Estimation and Inference for Causal Explainability
topic Methodology
Applications
url https://arxiv.org/abs/2512.20219