Uncertainty quantification for Bayesian nonparametric estimators of rare species variety

Species sampling problems have regained popularity in recent years due to their frequent appearance in challenging applications arising from ecology, genetics, linguistic, etc. Interest often lies in estimating the number of rare species that appear in a sample, that is species with a frequency smaller than a specific abundance threshold. The Bayesian nonparametric approach has proved successful by providing closed form estimators for rare species variety. In this paper we present a novel methodology for endowing such estimators with asymptotic credible intervals. We illustrate it through the analysis of some genomic datasets.