Kernel based Dirichlet sequences
Abstract
Let be a sequence of random variables with values in a standard space . Suppose \begin{gather*} X_1\sim\nu\quad\text{and}\quad P\bigl(X_{n+1}\in\cdot\mid X_1,\ldots,X_n\bigr)=\frac{\theta\nu(\cdot)+\sum_{i=1}^nK(X_i)(\cdot)}{n+\theta}\quad\quad\text{a.s.} \end{gather*} where is a constant, a probability measure on , and a random probability measure on . Then, is exchangeable whenever is a regular conditional distribution for given any sub--field of . Under this assumption, enjoys all the main properties of classical Dirichlet sequences, including Sethuraman's representation, conjugacy property, and convergence in total variation of predictive distributions. If is the weak limit of the empirical measures, conditions for to be a.s. discrete, or a.s. non-atomic, or a.s., are provided. Two CLT's are proved as well. The first deals with stable convergence while the second concerns total variation distance.
Cite
@article{arxiv.2106.00114,
title = {Kernel based Dirichlet sequences},
author = {Patrizia Berti and Emanuela Dreassi and Fabrizio Leisen and Luca Pratelli and Pietro Rigo},
journal= {arXiv preprint arXiv:2106.00114},
year = {2022}
}