EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Causal inference with misspecified network interference structure

Bar Weinstein iD, Daniel Nevo iD

Biometrics · 2026

DOI: 10.1093/biomtc/ujag023 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biometrics, 2026. Verfasst von Bar Weinstein und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Statistics and Operations Research, Tel Aviv University, Tel Aviv, 6997801, Israel. Hauptthemen laut MEDLINE: Causality, Models, Statistical. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Israel Science Foundation, Israeli Council For Higher Education Data Science Fellowship.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Under interference, the treatment of one unit may affect the outcomes of other units. Such interference patterns between units are typically represented by a network. Correctly specifying this network requires identifying which units can affect others—an inherently challenging task. Nevertheless, most existing approaches assume that a known and accurate network specification is given. In this paper, we study the consequences of such misspecification. We derive bounds on the bias arising from estimating causal effects using a misspecified network, showing that the estimation bias grows with the divergence between the assumed and true networks, quantified through their induced exposure probabilities. To address this challenge, we propose a novel estimator that leverages multiple networks simultaneously and remains unbiased if at least one of the networks is correct, even when we do not know which one. Therefore, the proposed estimator provides robustness to network specification. We illustrate key properties and demonstrate the utility of our proposed estimator through simulations and analysis of a social network field experiment.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

CausalityModels, StatisticalComputer SimulationBiasHumansData Interpretation, StatisticalBiometry

Stichwörter der Autor:innen

SUTVAexposure mappingmulti-layer networksnetwork experimentsspillovers

Autor:innen und Institutionen

  • Department of Statistics and Operations Research, Tel Aviv University, Tel Aviv, 6997801, Israel.

ORCID-Profile: Bar Weinstein, Daniel Nevo

Förderung

  • Israel Science Foundation
  • Israeli Council For Higher Education Data Science Fellowship

Zitieren

APA

Weinstein, B., & Nevo, D. (2026). Causal inference with misspecified network interference structure. Biometrics. https://doi.org/10.1093/biomtc/ujag023

Vancouver

Weinstein B, Nevo D. Causal inference with misspecified network interference structure. Biometrics. 2026. doi:10.1093/biomtc/ujag023

BibTeX
@article{Weinstein2026,
  author = {Weinstein, Bar and Nevo, Daniel},
  title = {Causal inference with misspecified network interference structure},
  journal = {Biometrics},
  year = {2026},
  doi = {10.1093/biomtc/ujag023}
}
Kostenlos starten →