Cover Image

Discrete random probability measures: a general framework for nonparametric Bayesian inference

A unifying framework for Bayesian analysis in discrete nonparametric settings is proposed. To this aim, a general class of nonparametric discrete prior distributions on an arbitrary sample space is introduced. The general structure of the posterior and predictive distributions and an explicit updating mechanism for the posterior are developed.
Year of publication:
2004
Authors: Ongaro, Andrea ; Cattaneo, Carla
Published in:
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 67.2004, 1, p. 33-45
Publisher: Elsevier
Keywords: Nonparametric priors Generalized Dirichlet process Mixture representation Random weights
Type of publication: Article
Source:
Persistent link: https://ebvufind01.dmz1.zbw.eu/10005319169
Similar items by person