Bayesian Model Averaging in Causal Instrumental Variable Models
Journal of Applied Econometrics · 2026-08 · 41 (5) · 669-683
DOI: 10.1002/jae.70070 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Applied Econometrics, online erschienen am 22.05.2026. Verfasst von Gregor Steiner und 1 weiteren Autor:innen. Die Arbeit nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Instrumental variables are a popular tool to infer causal effects under unobserved confounding, but choosing suitable instruments is challenging in practice. We propose gIVBMA, a Bayesian model averaging procedure that addresses this challenge by averaging across different sets of instrumental variables and covariates in a structural equation model. This allows for data‐driven selection of valid and relevant instruments and provides additional robustness against invalid instruments. Our approach extends previous work through a scale‐invariant prior structure and accommodates non‐Gaussian outcomes and treatments, offering greater flexibility than existing methods. The computational strategy uses conditional Bayes factors to update models separately for the outcome and treatments. We prove that this model selection procedure is consistent. In simulation experiments, gIVBMA outperforms current state‐of‐the‐art methods. We demonstrate its usefulness in two empirical applications: the effects of malaria and institutions on income per capita and the returns to schooling. A software implementation of gIVBMA is available in Julia.
Zitieren
APA
Steiner, G., & Steel, M. (2026). Bayesian Model Averaging in Causal Instrumental Variable Models. Journal of Applied Econometrics, 41(5), 669-683. https://doi.org/10.1002/jae.70070
Vancouver
Steiner G, Steel M. Bayesian Model Averaging in Causal Instrumental Variable Models. Journal of Applied Econometrics. 2026;41(5):669-683. doi:10.1002/jae.70070
BibTeX
@article{Steiner2026,
author = {Steiner, Gregor and Steel, Mark},
title = {Bayesian Model Averaging in Causal Instrumental Variable Models},
journal = {Journal of Applied Econometrics},
year = {2026},
volume = {41},
number = {5},
pages = {669--683},
doi = {10.1002/jae.70070}
}
EUVIMED