Genuinely Robust Inference for Clustered Data
Abstract
Conventional cluster-robust inference can be invalid when data contain clusters of unignorably large size. We formalize this issue by deriving a necessary and sufficient condition for its validity, and show that this condition is frequently violated in practice: specifications from 77% of empirical research articles in American Economic Review and Econometrica during 2020-2021 appear not to meet it. To address this limitation, we propose a genuinely robust inference procedure based on a new cluster score bootstrap. We establish its validity and size control across broad classes of data-generating processes where conventional methods break down. Simulation studies corroborate our theoretical findings, and empirical applications illustrate that employing the proposed method can substantially alter conventional statistical conclusions.
Keywords
Cite
@article{arxiv.2308.10138,
title = {Genuinely Robust Inference for Clustered Data},
author = {Harold D. Chiang and Yuya Sasaki and Yulong Wang},
journal= {arXiv preprint arXiv:2308.10138},
year = {2025}
}
Comments
This paper supersedes the manuscripts previously circulated under the titles "On the Inconsistency of Cluster-Robust Inference and How Subsampling Can Fix It" and "Non-Robustness of the Cluster-Robust Inference: with a Proposal of a New Robust Method" (arXiv:2210.16991)