English

Generalised Poisson-Dirichlet Distributions and the Negative Binomial Point Process

Probability 2017-03-30 v3

Abstract

When S=(St)t0S=(S_t)_{t\ge 0} is an α\alpha-stable subordinator, the sequence of ordered jumps of SS, up till time 11, omitting the rr largest of them, and taken as proportions of their sum (r)St^{(r)}S_t, defines a 2-parameter distribution on the infinite dimensional simplex, \nabla_{\infty}, which we call the PDα(r)\mathrm{PD}_\alpha^{(r)} distribution. When r=0r=0 it reduces to the PDα\mathrm{PD}_\alpha distribution introduced by Kingman in 1975. We observe a serendipitous connection between PDα(r)\mathrm{PD}_\alpha^{(r)} and the negative binomial point process of Gregoire (1984), which we exploit to analyse in detail a size-biased version of PDα(r)\mathrm{PD}_\alpha^{(r)}. As a consequence we derive a stick-breaking representation for the process and a useful form for its distribution. This program produces a large new class of distributions available for a variety of modelling purposes.

Keywords

Cite

@article{arxiv.1611.09980,
  title  = {Generalised Poisson-Dirichlet Distributions and the Negative Binomial Point Process},
  author = {Yuguang F. Ipsen and Ross A. Maller},
  journal= {arXiv preprint arXiv:1611.09980},
  year   = {2017}
}

Comments

17 pages

R2 v1 2026-06-22T17:08:55.258Z