Beta-product Poisson-Dirichlet Processes

e-Archivo Repository

Show simple item record Bassetti, Federico Casarin, Roberto Leisen, Fabrizio
dc.contributor.editor Universidad Carlos III de Madrid. Departamento de Estadística 2011-09-23T09:39:22Z 2011-09-23T09:39:22Z 2011-09
dc.description.abstract Time series data may exhibit clustering over time and, in a multiple time series context, the clustering behavior may differ across the series. This paper is motivated by the Bayesian non--parametric modeling of the dependence between the clustering structures and the distributions of different time series. We follow a Dirichlet process mixture approach and introduce a new class of multivariate dependent Dirichlet processes (DDP). The proposed DDP are represented in terms of vector of stick-breaking processes with dependent weights. The weights are beta random vectors that determine different and dependent clustering effects along the dimension of the DDP vector. We discuss some theoretical properties and provide an efficient Monte Carlo Markov Chain algorithm for posterior computation. The effectiveness of the method is illustrated with a simulation study and an application to the United States and the European Union industrial production indexes.
dc.format.mimetype application/pdf
dc.language.iso eng
dc.relation.ispartofseries UC3M Working papers. Statistics and Econometrics
dc.relation.ispartofseries 11-23
dc.rights Atribución-NoComercial-SinDerivadas 3.0 España
dc.subject.other Bayesian non--parametrics
dc.subject.other Dirichlet process
dc.subject.other Poisson-Dirichlet process
dc.subject.other Multiple Time-series non--parametrics
dc.title Beta-product Poisson-Dirichlet Processes
dc.type workingPaper
dc.subject.eciencia Estadística
dc.rights.accessRights openAccess
dc.type.version submitedVersion
dc.identifier.uxxi DT/0000000949
 Find Full text

Files in this item

*Click on file's image for preview. (Embargoed files's preview is not supported)

The following license files are associated with this item:

This item appears in the following Collection(s)

Show simple item record