Central limit theorems associated with the hierarchical Dirichlet process

TL;DR

This paper investigates the asymptotic behavior of the hierarchical Dirichlet process, establishing central limit theorems for its weight distributions as concentration parameters grow large, with implications across various fields.

Contribution

It provides the first central limit theorems for the power sum symmetric polynomials of hierarchical Dirichlet process weights, including explicit variance formulas highlighting the hierarchical structure's impact.

Findings

Established CLTs for the weights' power sums as concentration parameters increase

Derived explicit formulas for asymptotic variances showing hierarchical effects

Connected the results to measures in genetics, ecology, and economics

Abstract

The hierarchical Dirichlet process is a discrete random measure used as a prior in Bayesian nonparametrics and motivated by the study of groups of clustered data. We study the asymptotic behavior of the power sum symmetric polynomials for the vector of weights of the hierarchical Dirichlet process as the concentration parameters tend to infinity. We establish central limit theorems and obtain explicit representations for the asymptotic variances, with the latter clearly showing the impact of the hierarchical structure. These objects are related to the homozygosity in population genetics, the Simpson diversity index in ecology, and the Herfindahl-Hirschman index in economics.