Characterizations of Multivariate Normal-Poisson Model

Affiliation auteurs!!!! Error affiliation !!!!
TitreCharacterizations of Multivariate Normal-Poisson Model
Type de publicationJournal Article
Year of Publication2015
AuteursNisa K, Kokonendji CC, Saefuddin A
JournalJIRSS-JOURNAL OF THE IRANIAN STATISTICAL SOCIETY
Volume14
Pagination37-51
Type of ArticleArticle
ISSN1726-4057
Mots-clésGeneralized variance, Infinitely divisible measure, Monge-Ampere equation, Multivariate exponential family, Variance function
Résumé

Multivariate normal-Poisson model has been recently introduced as a special case of normal stable Tweedie models. The model is composed of a univariate Poisson variable, and the remaining variables given the Poisson one are independent Gaussian variables with variance the value of the Poisson component. Two characterizations of this model are shown, first by variance function and then by generalized variance function which is the determinant of the variance function. The latter provides an explicit solution of a particular Monge-Ampere equation.

DOI10.7508/jirss.2015.01.003