English

Copulas in three dimensions with prescribed correlations

Statistics Theory 2010-04-20 v1 Statistics Theory

Abstract

Given an arbitrary three-dimensional correlation matrix, we prove that there exists a three-dimensional joint distribution for the random variable (X,Y,Z)(X,Y,Z) such that XX,YY and ZZ are identically distributed with beta distribution βk,k(dx)\beta_{k,k}(dx) on (0,1)(0,1) if k1/2k\geq 1/2. This implies that any correlation structure can be attained for three-dimensional copulas.

Keywords

Cite

@article{arxiv.1004.3146,
  title  = {Copulas in three dimensions with prescribed correlations},
  author = {Luc Devroye and Gerard Letac},
  journal= {arXiv preprint arXiv:1004.3146},
  year   = {2010}
}

Comments

15 pages, 2 figures

R2 v1 2026-06-21T15:11:55.014Z