Saved in:
Bibliographic Details
Main Authors: Fu, Lianyan, Wang, Rui, Zhang, Zihan
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.29238
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914611996917760
author Fu, Lianyan
Wang, Rui
Zhang, Zihan
author_facet Fu, Lianyan
Wang, Rui
Zhang, Zihan
contents This paper proposes a generalized Mundlak estimator based on graph neural networks (GME-GNN). The estimator is designed to mitigate bias arising from group-level heterogeneity and to accommodate within-group dependence among individuals. Traditional fixed-effects models handle group heterogeneity via group-specific intercepts, but require overly strict linear additivity and intra-group independence assumptions, and are confined to within-group comparisons. Rather than relying on intercepts, GME-GNN uses aggregated group-level balancing statistics to fully control between-group confounding, enabling valid cross-group comparisons and relaxing linearity constraints. It further employs graph neural network message-passing to adaptively learn nonlinear representations and capture intra-group interaction effects. Theoretical analysis shows that the estimator satisfies double robustness and is asymptotically normal. Simulation and empirical studies confirm its performance.
format Preprint
id arxiv_https___arxiv_org_abs_2605_29238
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Graph Neural Networks for Generalized Mundlak Estimator under Network Confounding
Fu, Lianyan
Wang, Rui
Zhang, Zihan
Econometrics
This paper proposes a generalized Mundlak estimator based on graph neural networks (GME-GNN). The estimator is designed to mitigate bias arising from group-level heterogeneity and to accommodate within-group dependence among individuals. Traditional fixed-effects models handle group heterogeneity via group-specific intercepts, but require overly strict linear additivity and intra-group independence assumptions, and are confined to within-group comparisons. Rather than relying on intercepts, GME-GNN uses aggregated group-level balancing statistics to fully control between-group confounding, enabling valid cross-group comparisons and relaxing linearity constraints. It further employs graph neural network message-passing to adaptively learn nonlinear representations and capture intra-group interaction effects. Theoretical analysis shows that the estimator satisfies double robustness and is asymptotically normal. Simulation and empirical studies confirm its performance.
title Graph Neural Networks for Generalized Mundlak Estimator under Network Confounding
topic Econometrics
url https://arxiv.org/abs/2605.29238