The definition and investigation of general classes of non-parametric priors has recently been an active researchline in Bayesian statistics. Among the various proposals, the Gibbs-type family, which includes the Dirichlet process as a special case, stands out as the most tractable class of nonparametric priors for exchangeable sequences of observations. This is the consequence of a key simplifying assumption on the learning mechanism, which, however, has justification except that of ensuring mathematical tractability. In this paper, we remove such an assumption and investigate a general class of random probability measures going beyond the Gibs-type framework.More specifically, we present a nonparametric hierarchical structure based on transformations of completely random measures, which extends the popular hierarchical Dirichlet process. This class of priors preserves a good degree oftractability, given that we are able to determine the funda-mental quantities for Bayesian inference. In particular, we derive the induced partition structure and the prediction rules and characterize the posterior distribution. These theoretical results are also crucial to devise both a marginal and a conditional algorithm for posterior inference. An illustration concerning prediction in genomic sequencing is also provided.

Bayesian nonparametric inference beyond the Gibbs-type framework

Camerlenghi, Federico;Lijoi, Antonio;Pruenster, Igor
2018

Abstract

The definition and investigation of general classes of non-parametric priors has recently been an active researchline in Bayesian statistics. Among the various proposals, the Gibbs-type family, which includes the Dirichlet process as a special case, stands out as the most tractable class of nonparametric priors for exchangeable sequences of observations. This is the consequence of a key simplifying assumption on the learning mechanism, which, however, has justification except that of ensuring mathematical tractability. In this paper, we remove such an assumption and investigate a general class of random probability measures going beyond the Gibs-type framework.More specifically, we present a nonparametric hierarchical structure based on transformations of completely random measures, which extends the popular hierarchical Dirichlet process. This class of priors preserves a good degree oftractability, given that we are able to determine the funda-mental quantities for Bayesian inference. In particular, we derive the induced partition structure and the prediction rules and characterize the posterior distribution. These theoretical results are also crucial to devise both a marginal and a conditional algorithm for posterior inference. An illustration concerning prediction in genomic sequencing is also provided.
2018
Camerlenghi, Federico; Lijoi, Antonio; Pruenster, Igor
File in questo prodotto:
File Dimensione Formato  
Camerlenghi_et_al-2018-Scandinavian_Journal_of_Statistics.pdf

non disponibili

Tipologia: Pdf editoriale (Publisher's layout)
Licenza: NON PUBBLICO - Accesso privato/ristretto
Dimensione 808.67 kB
Formato Adobe PDF
808.67 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11565/4012459
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 6
  • ???jsp.display-item.citation.isi??? 5
social impact