Bayesian Additive Regression Trees for Modeling Multiple Exposures With Measurement Error
Statistics in Medicine · 2026
DOI: 10.1002/sim.70739 ↗Kurz gefasst
Fachartikel in der Zeitschrift Statistics in Medicine, 2026. Verfasst von Madeleine E St Ville und 4 weiteren Autor:innen. Beteiligte Einrichtung: Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland, USA. Hauptthemen laut MEDLINE: Models, Statistical. Die Arbeit nennt 43 Literaturangaben. Gefördert von: Intramural Research Program of the National Institutes of Health (NIH).
Abstract (Originalsprache)
We develop a Bayesian Additive Regression Trees with Measurement Error (BART‐ME) model for flexibly estimating exposure‐response functions when multiple covariates are measured with classical error. Unlike existing approaches, BART‐ME accommodates nonlinearities, interactions, correlated exposures, and correlated measurement errors, while exploiting replicate measurements to estimate the error variance‐covariance structure. Posterior inference is obtained via a Metropolis‐within‐Gibbs algorithm, yielding estimates of exposure‐response functions, variable importance, and measurement reliability. Simulation studies demonstrate that BART‐ME reduces bias and improves coverage compared with Naïve BART analyses that ignore measurement error, even with sparse replication. In an application to first‐trimester ultrasound data from the NICHD Fetal Growth Studies, BART‐ME identified crown‐rump length as the most reliable predictor of gestational age at delivery and revealed nonlinear patterns attenuated by Naïve methods. These results illustrate the potential of BART‐ME as a general framework for correcting measurement error in complex exposure settings.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland, USA.
- Department of Applied Statistics, Chung-Ang University, Seoul, Korea.
- Institute for Informatics, Data Science, and Biostatistics, Washington University School of Medicine, St. Louis, MO, USA.
ORCID-Profile: Madeleine E St Ville, Yaeji Lim, Ruijin Lu, Katherine L Grantz
Förderung
- Intramural Research Program of the National Institutes of Health (NIH)
Zitieren
APA
St Ville, M. E., Lim, Y., Lu, R., Grantz, K. L., & Chen, Z. (2026). Bayesian Additive Regression Trees for Modeling Multiple Exposures With Measurement Error. Statistics in Medicine. https://doi.org/10.1002/sim.70739
Vancouver
St Ville ME, Lim Y, Lu R, Grantz KL, Chen Z. Bayesian Additive Regression Trees for Modeling Multiple Exposures With Measurement Error. Statistics in Medicine. 2026. doi:10.1002/sim.70739
BibTeX
@article{StVille2026,
author = {St Ville, Madeleine E and Lim, Yaeji and Lu, Ruijin and Grantz, Katherine L and Chen, Zhen},
title = {Bayesian Additive Regression Trees for Modeling Multiple Exposures With Measurement Error},
journal = {Statistics in Medicine},
year = {2026},
doi = {10.1002/sim.70739}
}
EUVIMED