English

Characterising random partitions by random colouring

Probability 2020-01-14 v2

Abstract

Let (X1,X2,...)(X_1,X_2,...) be a random partition of the unit interval [0,1][0,1], i.e. Xi0X_i\geq0 and i1Xi=1\sum_{i\geq1} X_i=1, and let (ε1,ε2,...)(\varepsilon_1,\varepsilon_2,...) be i.i.d. Bernoulli random variables of parameter p(0,1)p \in (0,1). The Bernoulli convolution of the partition is the random variable Z=i1εiXiZ =\sum_{i\geq1} \varepsilon_i X_i. The question addressed in this article is: Knowing the distribution of ZZ for some fixed p(0,1)p\in(0,1), what can we infer about the random partition? We consider random partitions formed by residual allocation and prove that their distributions are fully characterised by their Bernoulli convolution if and only if the parameter pp is not equal to 1/21/2.

Keywords

Cite

@article{arxiv.1907.05960,
  title  = {Characterising random partitions by random colouring},
  author = {Jakob E. Björnberg and Cécile Mailler and Peter Mörters and Daniel Ueltschi},
  journal= {arXiv preprint arXiv:1907.05960},
  year   = {2020}
}

Comments

12 pages

R2 v1 2026-06-23T10:20:01.615Z