Bayesian hierarchical modeling based on multisource exchangeability

Alexander M. Kaizer, Joe Koopmeiners, Brian P. Hobbs

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Bayesian hierarchical models produce shrinkage estimators that can be used as the basis for integrating supplementary data into the analysis of a primary data source. Established approaches should be considered limited, however, because posterior estimation either requires prespecification of a shrinkage weight for each source or relies on the data to inform a single parameter, which determines the extent of influence or shrinkage from all sources, risking considerable bias or minimal borrowing.We introduce multisource exchangeability models (MEMs), a general Bayesian approach for integrating multiple, potentially nonexchangeable, supplemental data sources into the analysis of a primary data source. Our proposed modeling framework yields source-specific smoothing parameters that can be estimated in the presence of the data to facilitate a dynamic multi-resolution smoothed estimator that is asymptotically consistent while reducing the dimensionality of the prior space. When compared with competing Bayesian hierarchical modeling strategies, we demonstrate that MEMs achieve approximately 2.2 times larger median effective supplemental sample size when the supplemental data sources are exchangeable as well as a 56% reduction in bias when there is heterogeneity among the supplemental sources.We illustrate the application ofMEMs using a recently completed randomized trial of very low nicotine content cigarettes, which resulted in a 30% improvement in efficiency compared with the standard analysis.

Original languageEnglish (US)
Pages (from-to)169-184
Number of pages16
JournalBiostatistics
Volume19
Issue number2
DOIs
StatePublished - Apr 1 2018

Fingerprint

Exchangeability
Hierarchical Modeling
Bayesian Modeling
Shrinkage
Effective Sample Size
Bayesian Hierarchical Model
Shrinkage Estimator
Randomized Trial
Modeling
Data sources
Smoothing Parameter
Multiresolution
Bayesian Approach
Dimensionality
Estimator

Keywords

  • Bayesian hierarchical modeling
  • Heterogeneous sources of data
  • Multisource smoothing
  • Supplementary data

PubMed: MeSH publication types

  • Journal Article
  • Research Support, N.I.H., Extramural
  • Research Support, U.S. Gov't, P.H.S.

Cite this

Bayesian hierarchical modeling based on multisource exchangeability. / Kaizer, Alexander M.; Koopmeiners, Joe; Hobbs, Brian P.

In: Biostatistics, Vol. 19, No. 2, 01.04.2018, p. 169-184.

Research output: Contribution to journalArticle

Kaizer, Alexander M. ; Koopmeiners, Joe ; Hobbs, Brian P. / Bayesian hierarchical modeling based on multisource exchangeability. In: Biostatistics. 2018 ; Vol. 19, No. 2. pp. 169-184.
@article{e17e9177f12645d382d258f72ee3a063,
title = "Bayesian hierarchical modeling based on multisource exchangeability",
abstract = "Bayesian hierarchical models produce shrinkage estimators that can be used as the basis for integrating supplementary data into the analysis of a primary data source. Established approaches should be considered limited, however, because posterior estimation either requires prespecification of a shrinkage weight for each source or relies on the data to inform a single parameter, which determines the extent of influence or shrinkage from all sources, risking considerable bias or minimal borrowing.We introduce multisource exchangeability models (MEMs), a general Bayesian approach for integrating multiple, potentially nonexchangeable, supplemental data sources into the analysis of a primary data source. Our proposed modeling framework yields source-specific smoothing parameters that can be estimated in the presence of the data to facilitate a dynamic multi-resolution smoothed estimator that is asymptotically consistent while reducing the dimensionality of the prior space. When compared with competing Bayesian hierarchical modeling strategies, we demonstrate that MEMs achieve approximately 2.2 times larger median effective supplemental sample size when the supplemental data sources are exchangeable as well as a 56{\%} reduction in bias when there is heterogeneity among the supplemental sources.We illustrate the application ofMEMs using a recently completed randomized trial of very low nicotine content cigarettes, which resulted in a 30{\%} improvement in efficiency compared with the standard analysis.",
keywords = "Bayesian hierarchical modeling, Heterogeneous sources of data, Multisource smoothing, Supplementary data",
author = "Kaizer, {Alexander M.} and Joe Koopmeiners and Hobbs, {Brian P.}",
year = "2018",
month = "4",
day = "1",
doi = "10.1093/biostatistics/kxx031",
language = "English (US)",
volume = "19",
pages = "169--184",
journal = "Biostatistics",
issn = "1465-4644",
publisher = "Oxford University Press",
number = "2",

}

TY - JOUR

T1 - Bayesian hierarchical modeling based on multisource exchangeability

AU - Kaizer, Alexander M.

AU - Koopmeiners, Joe

AU - Hobbs, Brian P.

PY - 2018/4/1

Y1 - 2018/4/1

N2 - Bayesian hierarchical models produce shrinkage estimators that can be used as the basis for integrating supplementary data into the analysis of a primary data source. Established approaches should be considered limited, however, because posterior estimation either requires prespecification of a shrinkage weight for each source or relies on the data to inform a single parameter, which determines the extent of influence or shrinkage from all sources, risking considerable bias or minimal borrowing.We introduce multisource exchangeability models (MEMs), a general Bayesian approach for integrating multiple, potentially nonexchangeable, supplemental data sources into the analysis of a primary data source. Our proposed modeling framework yields source-specific smoothing parameters that can be estimated in the presence of the data to facilitate a dynamic multi-resolution smoothed estimator that is asymptotically consistent while reducing the dimensionality of the prior space. When compared with competing Bayesian hierarchical modeling strategies, we demonstrate that MEMs achieve approximately 2.2 times larger median effective supplemental sample size when the supplemental data sources are exchangeable as well as a 56% reduction in bias when there is heterogeneity among the supplemental sources.We illustrate the application ofMEMs using a recently completed randomized trial of very low nicotine content cigarettes, which resulted in a 30% improvement in efficiency compared with the standard analysis.

AB - Bayesian hierarchical models produce shrinkage estimators that can be used as the basis for integrating supplementary data into the analysis of a primary data source. Established approaches should be considered limited, however, because posterior estimation either requires prespecification of a shrinkage weight for each source or relies on the data to inform a single parameter, which determines the extent of influence or shrinkage from all sources, risking considerable bias or minimal borrowing.We introduce multisource exchangeability models (MEMs), a general Bayesian approach for integrating multiple, potentially nonexchangeable, supplemental data sources into the analysis of a primary data source. Our proposed modeling framework yields source-specific smoothing parameters that can be estimated in the presence of the data to facilitate a dynamic multi-resolution smoothed estimator that is asymptotically consistent while reducing the dimensionality of the prior space. When compared with competing Bayesian hierarchical modeling strategies, we demonstrate that MEMs achieve approximately 2.2 times larger median effective supplemental sample size when the supplemental data sources are exchangeable as well as a 56% reduction in bias when there is heterogeneity among the supplemental sources.We illustrate the application ofMEMs using a recently completed randomized trial of very low nicotine content cigarettes, which resulted in a 30% improvement in efficiency compared with the standard analysis.

KW - Bayesian hierarchical modeling

KW - Heterogeneous sources of data

KW - Multisource smoothing

KW - Supplementary data

UR - http://www.scopus.com/inward/record.url?scp=85044738207&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044738207&partnerID=8YFLogxK

U2 - 10.1093/biostatistics/kxx031

DO - 10.1093/biostatistics/kxx031

M3 - Article

VL - 19

SP - 169

EP - 184

JO - Biostatistics

JF - Biostatistics

SN - 1465-4644

IS - 2

ER -