Sums of Possibly Associated Bernoulli Variables: The Conway-Maxwell-Binomial Distribution
Methodology
2014-04-08 v1
Abstract
The study of sums of possibly associated Bernoulli random variables has been hampered by an asymmetry between positive correlation and negative correlation. The Conway-Maxwell Binomial (COMB) distribution and its multivariate extension, the Conway-Maxwell Multinomial (COMM) distribution, gracefully model both positive and negative association. Sufficient statistics and a family of proper conjugate distributions are found. The relationship of this distribution to the exchangeable special case is explored, and two applications are discussed.
Cite
@article{arxiv.1404.1856,
title = {Sums of Possibly Associated Bernoulli Variables: The Conway-Maxwell-Binomial Distribution},
author = {Joseph B. Kadane},
journal= {arXiv preprint arXiv:1404.1856},
year = {2014}
}
Comments
17 pages, 4 figures