Approximating predictive probabilities of Gibbs-type priors - Archive ouverte HAL Access content directly
Journal Articles Sankhya A Year : 2021

Approximating predictive probabilities of Gibbs-type priors


Gibbs-type random probability measures, or Gibbs-type priors, are arguably the most “natural” generalization of the celebrated Dirichlet prior. Among them the two parameter Poisson–Dirichlet prior certainly stands out in terms of mathematical tractability and interpretability of its predictive probabilities, which made it the natural candidate in a plethora of applications. Given a random sample of size n from an arbitrary Gibbs-type prior, we show that the corresponding predictive probabilities admit a large n approximation, with an error term vanishing as o(1/n), which maintains the same desirable features as the predictive probabilities of the two parameter Poisson–Dirichlet prior. Our result is illustrated through an extensive simulation study, which includes an application in the context of Bayesian nonparametric mixture modeling.
Fichier principal
Vignette du fichier
arbel-favaro.pdf (652.84 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01693333 , version 1 (26-01-2018)



Julyan Arbel, Stefano Favaro. Approximating predictive probabilities of Gibbs-type priors. Sankhya A, 2021, 83, pp.496-519. ⟨10.1007/s13171-019-00187-y⟩. ⟨hal-01693333⟩
335 View
215 Download



Gmail Facebook X LinkedIn More