Open Access
2014 On the stick-breaking representation of $\sigma$-stable Poisson-Kingman models
Stefano Favaro, Maria Lomeli, Bernardo Nipoti, Yee Whye Teh
Electron. J. Statist. 8(1): 1063-1085 (2014). DOI: 10.1214/14-EJS921

Abstract

In this paper we investigate the stick-breaking representation for the class of $\sigma$-stable Poisson-Kingman models, also known as Gibbs-type random probability measures. This class includes as special cases most of the discrete priors commonly used in Bayesian nonparametrics, such as the two parameter Poisson-Dirichlet process and the normalized generalized Gamma process. Under the assumption $\sigma=u/v$, for any coprime integers $1\leq u<v$ such that $u/v\leq1/2$, we show that a $\sigma$-stable Poisson-Kingman model admits an explicit stick-breaking representation in terms of random variables which are obtained by suitably transforming Gamma random variables and products of independent Beta and Gamma random variables.

Citation

Download Citation

Stefano Favaro. Maria Lomeli. Bernardo Nipoti. Yee Whye Teh. "On the stick-breaking representation of $\sigma$-stable Poisson-Kingman models." Electron. J. Statist. 8 (1) 1063 - 1085, 2014. https://doi.org/10.1214/14-EJS921

Information

Published: 2014
First available in Project Euclid: 5 August 2014

zbMATH: 1298.62049
MathSciNet: MR3263112
Digital Object Identifier: 10.1214/14-EJS921

Subjects:
Primary: 60G57 , 62F15

Keywords: $\sigma$-stable Poisson-Kingman model , $G$-functions , Bayesian nonparametrics , Beta random variable , discrete random probability measure , exponential tilting , Gamma random variable , stick-breaking prior

Rights: Copyright © 2014 The Institute of Mathematical Statistics and the Bernoulli Society

Vol.8 • No. 1 • 2014
Back to Top