EUVIMEDObjevte EUVIMED
Open access · CC BY 4.0

Bayesian Estimation of Hierarchical Linear Models From Incomplete Data: Cluster‐Level Interaction Effects and Small Sample Sizes

Dongho Shin iD, Yongyun Shin iD, Nao Hagiwara iD

Statistics in Medicine · 2025

DOI: 10.1002/sim.70051 ↗

At a glance

Journal article in Statistics in Medicine, 2025. Written by Dongho Shin and 2 co-authors. Institution: Department of Biostatistics, Virginia Commonwealth University, Virginia, USA. 36 references. The full text is freely available (open access, CC BY 4.0). Funded by: NCI NIH HHS, NINR NIH HHS, Institute of Education Sciences.

AbstraktSubjectsBibliografické údajeCiteRelated studies

Abstract (original language)

We consider Bayesian estimation of a hierarchical linear model (HLM) from partially observed data, assumed to be missing at random, and small sample sizes. A vector of continuous covariates includes cluster‐level partially observed covariates with interaction effects. Due to small sample sizes from 37 patient–physician encounters repeatedly measured at four time points, maximum‐likelihood estimation is suboptimal. Existing Gibbs samplers impute missing values of by a Metropolis algorithm using proposal densities that have constant variances while the target posterior distributions have nonconstant variances. Therefore, these samplers may not ensure compatibility with the HLM and, as a result, may not guarantee unbiased estimation of the HLM. We introduce a compatible Gibbs sampler that imputes parameters and missing values directly from the exact posterior distributions. We apply our Gibbs sampler to the longitudinal patient–physician encounter data and compare our estimators with those from existing methods by simulation.

Subjects

MeSH terms (MEDLINE)

Bayes TheoremHumansLinear ModelsSample SizeComputer SimulationCluster AnalysisAlgorithmsLikelihood FunctionsLongitudinal Studies

Author keywords

Compatible Gibbs samplerExact posterior distributionsInteraction effects of continuous covariatesProvisionally known random effectsRobust estimation

Authors and institutions

  • Department of Biostatistics, Virginia Commonwealth University, Virginia, USA.
  • Tempus AI, Inc., Chicago, Illinois, USA.
  • Department of Public Health Sciences, University of Virginia, Virginia, USA.

ORCID profiles: Dongho Shin, Yongyun Shin, Nao Hagiwara

Funding

  • NCI NIH HHS
  • NINR NIH HHS
  • Institute of Education Sciences
  • NIDDK NIH HHS
  • National Institute of Diabetes and Digestive and Kidney Diseases
  • National Institute of Nursing Research

Cite

APA

Shin, D., Shin, Y., & Hagiwara, N. (2025). Bayesian Estimation of Hierarchical Linear Models From Incomplete Data: Cluster‐Level Interaction Effects and Small Sample Sizes. Statistics in Medicine. https://doi.org/10.1002/sim.70051

Vancouver

Shin D, Shin Y, Hagiwara N. Bayesian Estimation of Hierarchical Linear Models From Incomplete Data: Cluster‐Level Interaction Effects and Small Sample Sizes. Statistics in Medicine. 2025. doi:10.1002/sim.70051

BibTeX
@article{Shin2025,
  author = {Shin, Dongho and Shin, Yongyun and Hagiwara, Nao},
  title = {Bayesian Estimation of Hierarchical Linear Models From Incomplete Data: Cluster‐Level Interaction Effects and Small Sample Sizes},
  journal = {Statistics in Medicine},
  year = {2025},
  doi = {10.1002/sim.70051}
}
Start for free →