Functionals of random probability measures are probabilistic objects whose properties are studied in different fields. They also play an important role in Bayesian Nonparametrics: understanding the behavior of a finite dimensional feature of a flexible and infinite-dimensional prior is crucial for prior elicitation. In particular distributions of means of nonparametric priors have been the object of thorough investigation in the literature. We target the inverse path: the determination of the parameter measure of a random probability measure giving rise to a fixed mean distribution. This direction yields a better understanding of the sets of mean distributions of notable nonparametric priors, giving moreover a way to directly enforce prior information, without losing inferential power. Here we summarize and report results obtained in [6] for the Dirichlet process, the normalized stable random measure and the Pitman–Yor process, with an application to mixture models.

Specification of the base measure of nonparametric priors via random means

Gaffi, Francesco
;
Lijoi, Antonio;Pruenster, Igor
2022

Abstract

Functionals of random probability measures are probabilistic objects whose properties are studied in different fields. They also play an important role in Bayesian Nonparametrics: understanding the behavior of a finite dimensional feature of a flexible and infinite-dimensional prior is crucial for prior elicitation. In particular distributions of means of nonparametric priors have been the object of thorough investigation in the literature. We target the inverse path: the determination of the parameter measure of a random probability measure giving rise to a fixed mean distribution. This direction yields a better understanding of the sets of mean distributions of notable nonparametric priors, giving moreover a way to directly enforce prior information, without losing inferential power. Here we summarize and report results obtained in [6] for the Dirichlet process, the normalized stable random measure and the Pitman–Yor process, with an application to mixture models.
2022
9783031164262
9783031164279
Argiento, Raffaele; Camerlenghi, Federico; Paganin, Sally
New frontiers in Bayesian statistics : BAYSM 2021, online, September 1-3
Gaffi, Francesco; Lijoi, Antonio; Pruenster, Igor
File in questo prodotto:
File Dimensione Formato  
BAYSM_published.pdf

non disponibili

Descrizione: paper
Tipologia: Pdf editoriale (Publisher's layout)
Licenza: NON PUBBLICO - Accesso privato/ristretto
Dimensione 185.31 kB
Formato Adobe PDF
185.31 kB Adobe PDF   Visualizza/Apri

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: https://hdl.handle.net/11565/4051866
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 0
social impact