Decompounding discrete distributions: A nonparametric Bayesian approach

DOI10.1111/SJOS.12413zbMATH Open1450.62029arXiv1903.11142OpenAlexW3103352383WikidataQ126988067 ScholiaQ126988067MaRDI QIDQ5118466FDOQ5118466

Shota Gugushvili, Frank van der Meulen, Ester Mariucci

Publication date: 8 September 2020

Published in: Scandinavian Journal of Statistics (Search for Journal in Brave)

Abstract: Suppose that a compound Poisson process is observed discretely in time and assume that its jump distribution is supported on the set of natural numbers. In this paper we propose a non-parametric Bayesian approach to estimate the intensity of the underlying Poisson process and the distribution of the jumps. We provide a MCMC scheme for obtaining samples from the posterior. We apply our method on both simulated and real data examples, and compare its performance with the frequentist plug-in estimator proposed by Buchmann and Gr"ubel. On a theoretical side, we study the posterior from the frequentist point of view and prove that as the sample size $nightarrowinfty$, it contracts around the `true', data-generating parameters at rate $1/sqrtn$, up to a $logn$ factor.

Full work available at URL: https://arxiv.org/abs/1903.11142

Recommendations

• Bayesian analysis for mixtures of discrete distributions with a non-parametric component
• Discrete random probability measures: a general framework for nonparametric Bayesian in\-fer\-ence
• Discrete approximation of multivariate densities with application to Bayesian estimation
• Compound random measures and their use in Bayesian non-parametrics
• On nonparametric Bayesian inference for the distribution of a random sample
• An asymptotic analysis of a class of discrete nonparametric priors
• Bayesian Nonparametric Inference for Random Distributions and Related Functions

zbMATH Keywords

data augmentationGibbs samplerMetropolis-Hastings algorithmDiophantine equationnonparametric Bayesian estimationcompound Poisson process

Mathematics Subject Classification ID

Density estimation (62G07) Point processes (e.g., Poisson, Cox, Hawkes processes) (60G55)

Cited In (5)

• Decompounding Poisson random sums: recursively truncated estimates in the discrete case
• Efficient approximation of probability distributions with \(k\)-order decomposable models
• Discretizing a compound distribution with application to categorical modelling
• Nonparametric Bayesian volatility estimation for gamma-driven stochastic differential equations
• Decompounding: an estimation problem for Poisson random sums.

This page was built for publication: Decompounding discrete distributions: A nonparametric Bayesian approach

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5118466)