EUVIMEDEUVIMED entdecken
Open Access

BAMITA: Bayesian multiple imputation for tensor arrays

Ziren Jiang, Gen Li iD, Eric F Lock iD

Biostatistics · 2025

DOI: 10.1093/biostatistics/kxae047 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biostatistics, 2025. Verfasst von Ziren Jiang und 2 weiteren Autor:innen. Beteiligte Einrichtung: Division of Biostatistics and Health Data Science, School of Public Health, University of Minnesota, 2221 University Avenue SE, Minneapolis, MN 55414…. Hauptthemen laut MEDLINE: Models, Statistical. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 24 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIGMS NIH HHS, NHGRI NIH HHS, NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Data increasingly take the form of a multi-way array, or tensor, in several biomedical domains. Such tensors are often incompletely observed. For example, we are motivated by longitudinal microbiome studies in which several timepoints are missing for several subjects. There is a growing literature on missing data imputation for tensors. However, existing methods give a point estimate for missing values without capturing uncertainty. We propose a multiple imputation approach for tensors in a flexible Bayesian framework, that yields realistic simulated values for missing entries and can propagate uncertainty through subsequent analyses. Our model uses efficient and widely applicable conjugate priors for a CANDECOMP/PARAFAC (CP) factorization, with a separable residual covariance structure. This approach is shown to perform well with respect to both imputation accuracy and uncertainty calibration, for scenarios in which either single entries or entire fibers of the tensor are missing. For two microbiome applications, it is shown to accurately capture uncertainty in the full microbiome profile at missing timepoints and used to infer trends in species diversity for the population. Documented R code to perform our multiple imputation approach is available at https://github.com/lockEF/MultiwayImputation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Models, StatisticalBayes TheoremHumansMicrobiotaData Interpretation, StatisticalUncertaintyBiostatistics

Stichwörter der Autor:innen

Bayesian inferencemicrobiome datamissing datamultiple imputationmultiway data

Autor:innen und Institutionen

  • Division of Biostatistics and Health Data Science, School of Public Health, University of Minnesota, 2221 University Avenue SE, Minneapolis, MN 55414, United States.
  • Department of Biostatistics, School of Public Health, University of Michigan, 1415 Washington Heights, M4210, Ann Arbor, MI 48109, United States.

ORCID-Profile: Gen Li, Eric F Lock

Förderung

  • NIGMS NIH HHS
  • NHGRI NIH HHS
  • NIH HHS
  • National Institutes of Health

Zitieren

APA

Jiang, Z., Li, G., & Lock, E. F. (2025). BAMITA: Bayesian multiple imputation for tensor arrays. Biostatistics. https://doi.org/10.1093/biostatistics/kxae047

Vancouver

Jiang Z, Li G, Lock EF. BAMITA: Bayesian multiple imputation for tensor arrays. Biostatistics. 2025. doi:10.1093/biostatistics/kxae047

BibTeX
@article{Jiang2025,
  author = {Jiang, Ziren and Li, Gen and Lock, Eric F},
  title = {BAMITA: Bayesian multiple imputation for tensor arrays},
  journal = {Biostatistics},
  year = {2025},
  doi = {10.1093/biostatistics/kxae047}
}
Kostenlos starten →