Bayesian nonparametric analysis for a species sampling model with finitely many types

2010/01/01 by Annalisa Cerquetti, Cerquetti, Annalisa
Computer Science · Mathematics · #60G09 #60G58 #Bayesian Methods and Mixture Models #Census and Population Estimation #FOS: Mathematics #Probability (math.PR) #Statistical Methods and Bayesian Inference #Statistics Theory (math.ST)

paper · pdf · doi:10.48550/arxiv.1001.0245

openalex publication_date 2010/01/01 · openalex created_date 2016/06/24 · openalex updated_date 2026/07/28

Abstract

We derive explicit Bayesian nonparametric analysis for a species sampling model with finitely many types of Gibbs form of type α= -1 recently introduced in Gnedin (2009). Our results complement existing analysis under Gibbs priors of type α∈ [0, 1) proposed in Lijoi et al. (2008). Calculations rely on a groups sequential construction of Gibbs partitions introduced in Cerquetti (2008).

Related

• Yet another application of marginals of multivariate Gibbs distributions
• Bayesian nonparametric estimation of Simpson's evenness index under α-Gibbs priors
• Posterior distributions of Gibbs-type priors
• Bayesian Nonparametric Inference for "Species-sampling" Problems
• A new parametrization of the Gnedin-Fisher species sampling model