A blocked Gibbs sampler for NGG-mixture models via a priori truncation (Articolo in rivista)

Type
Label
  • A blocked Gibbs sampler for NGG-mixture models via a priori truncation (Articolo in rivista) (literal)
Anno
  • 2015-01-01T00:00:00+01:00 (literal)
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#doi
  • 10.1007/s11222-015-9549-6 (literal)
Alternative label
  • Argiento R., Bianchini, I., Guglielmi, A. (2015)
    A blocked Gibbs sampler for NGG-mixture models via a priori truncation
    in Statistics and computing (Dordr., Online)
    (literal)
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#autori
  • Argiento R., Bianchini, I., Guglielmi, A. (literal)
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#url
  • http://link.springer.com/article/10.1007%2Fs11222-015-9549-6 (literal)
Rivista
Note
  • Scopu (literal)
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#affiliazioni
  • CNR-IMATI, CNR-IMATI, Politecnico di Milano (literal)
Titolo
  • A blocked Gibbs sampler for NGG-mixture models via a priori truncation (literal)
Abstract
  • We define a new class of random probability measures, approximating the well-known normalized gen- eralized gamma (NGG) process. Our new process is defined from the representation of NGG processes as discrete mea- sures where the weights are obtained by normalization of the jumps of a Poisson process, and the support consists of independent identically distributed location points, how- ever considering only jumps larger than a threshold ?. There- fore, the number of jumps of the new process, called ?-NGG process, is a.s. finite. A prior distribution for ? can be elicited. We assume such a process as the mixing measure in a mix- ture model for density and cluster estimation, and build an efficient Gibbs sampler scheme to simulate from the pos- terior. Finally, we discuss applications and performance of the model to two popular datasets, as well as comparison with competitor algorithms, the slice sampler and a posteri- ori truncation. (literal)
Prodotto di
Autore CNR
Insieme di parole chiave

Incoming links:


Prodotto
Autore CNR di
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#rivistaDi
Insieme di parole chiave di
data.CNR.it