Please use this identifier to cite or link to this item: http://buratest.brunel.ac.uk/handle/2438/9505
Full metadata record
DC FieldValueLanguage
dc.contributor.authorHastie, DI-
dc.contributor.authorLiverani, S-
dc.contributor.authorRichardson, S-
dc.date.accessioned2014-12-12T15:28:58Z-
dc.date.available2014-12-12T15:28:58Z-
dc.date.issued2014-
dc.identifier.citationStatistics and Computing, 2014en_US
dc.identifier.issn0960-3174-
dc.identifier.urihttp://link.springer.com/article/10.1007%2Fs11222-014-9471-3-
dc.identifier.urihttp://bura.brunel.ac.uk/handle/2438/9505-
dc.description.abstractWe consider the question of Markov chain Monte Carlo sampling from a general stick-breaking Dirichlet process mixture model, with concentration parameter (Formula presented.). This paper introduces a Gibbs sampling algorithm that combines the slice sampling approach of Walker (Communications in Statistics - Simulation and Computation 36:45-54, 2007) and the retrospective sampling approach of Papaspiliopoulos and Roberts (Biometrika 95(1):169-186, 2008). Our general algorithm is implemented as efficient open source C++ software, available as an R package, and is based on a blocking strategy similar to that suggested by Papaspiliopoulos (A note on posterior sampling from Dirichlet mixture models, 2008) and implemented by Yau et al. (Journal of the Royal Statistical Society, Series B (Statistical Methodology) 73:37-57, 2011). We discuss the difficulties of achieving good mixing in MCMC samplers of this nature in large data sets and investigate sensitivity to initialisation. We additionally consider the challenges when an additional layer of hierarchy is added such that joint inference is to be made on (Formula presented.). We introduce a new label-switching move and compute the marginal partition posterior to help to surmount these difficulties. Our work is illustrated using a profile regression (Molitor et al. Biostatistics 11(3):484-498, 2010) application, where we demonstrate good mixing behaviour for both synthetic and real examples. © 2014 The Author(s).en_US
dc.languageeng-
dc.language.isoenen_US
dc.publisherSpringer USen_US
dc.subjectBayesian clusteringen_US
dc.subjectDirichlet processen_US
dc.subjectMixture modelen_US
dc.subjectProfile regressionen_US
dc.titleSampling from Dirichlet process mixture models with unknown concentration parameter: mixing issues in large data implementationsen_US
dc.typeArticleen_US
dc.identifier.doihttp://dx.doi.org/10.1007/s11222-014-9471-3-
pubs.organisational-data/Brunel-
pubs.organisational-data/Brunel/Brunel Staff by College/Department/Division-
pubs.organisational-data/Brunel/Brunel Staff by College/Department/Division/College of Engineering, Design and Physical Sciences-
pubs.organisational-data/Brunel/Brunel Staff by College/Department/Division/College of Engineering, Design and Physical Sciences/Dept of Mathematics-
pubs.organisational-data/Brunel/Brunel Staff by College/Department/Division/College of Engineering, Design and Physical Sciences/Dept of Mathematics/Mathematical Sciences-
Appears in Collections:Dept of Computer Science Research Papers

Files in This Item:
File Description SizeFormat 
Fullpaper.pdf661.41 kBAdobe PDFView/Open


Items in BURA are protected by copyright, with all rights reserved, unless otherwise indicated.