Show simple item record

dc.contributor.authorManderson, Andrew
dc.contributor.authorGoudie, Robert
dc.date.accessioned2022-03-17T10:04:20Z
dc.date.available2022-03-17T10:04:20Z
dc.date.issued2022-04-15
dc.date.submitted2020-12-09
dc.identifier.issn0960-3174
dc.identifier.others11222-022-10086-2
dc.identifier.other10086
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/335090
dc.description.abstract<jats:title>Abstract</jats:title><jats:p>When statistical analyses consider multiple data sources, Markov melding provides a method for combining the source-specific Bayesian models. Markov melding joins together submodels that have a common quantity. One challenge is that the prior for this quantity can be implicit, and its prior density must be estimated. We show that error in this density estimate makes the two-stage Markov chain Monte Carlo sampler employed by Markov melding unstable and unreliable. We propose a robust two-stage algorithm that estimates the required prior marginal self-density ratios using weighted samples, dramatically improving accuracy in the tails of the distribution. The stabilised version of the algorithm is pragmatic and provides reliable inference. We demonstrate our approach using an evidence synthesis for inferring HIV prevalence, and an evidence synthesis of A/H1N1 influenza.</jats:p>
dc.languageen
dc.publisherSpringer Science and Business Media LLC
dc.subjectArticle
dc.subjectBiased sampling
dc.subjectData integration
dc.subjectEvidence synthesis
dc.subjectKernel density estimation
dc.subjectMulti-source inference
dc.subjectSelf-density ratio
dc.subjectWeighted sampling
dc.titleA numerically stable algorithm for integrating Bayesian models using Markov melding
dc.typeArticle
dc.date.updated2022-03-17T10:04:20Z
prism.issueIdentifier2
prism.publicationNameStatistics and Computing
prism.volume32
dc.identifier.doi10.17863/CAM.82532
dcterms.dateAccepted2022-01-29
rioxxterms.versionofrecord10.1007/s11222-022-10086-2
rioxxterms.versionVoR
rioxxterms.licenseref.urihttp://creativecommons.org/licenses/by/4.0/
dc.contributor.orcidManderson, Andrew [0000-0002-4946-9016]
dc.contributor.orcidGoudie, Robert [0000-0001-9554-1499]
dc.identifier.eissn1573-1375
pubs.funder-project-idAlan Turing Institute (EP/N510129/1)
pubs.funder-project-idMedical Research Council (GB) (MC_UU_00002/2)
cam.issuedOnline2022-02-18


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record