On a Multiplicative Multivariate Gamma Distribution With Applications in Insurance

25 Pages Posted: 14 Feb 2018

See all articles by Vadim Semenikhine

Vadim Semenikhine

York University

Edward Furman

York University - Department of Mathematics and Statistics

Jianxi Su

Purdue University - Department of Statistics

Date Written: February 2, 2018

Abstract

In a recent paper [Albrecher, Constantinescu and Loisel (2011). Explicit ruin formulas for models with dependence among risks. Insurance: Mathematics and Economics 48(2), 265 – 270] Professors Hansjörg Albrecher, Corina Constantinescu and Stephane Loisel noted – in passing – a way to employ exponential mixtures for formulating multivariate probability distributions with dependent univariate margins distributed gamma. The main message of our report is that the probabilistic construction in ibid., which has been arguably overlooked by the actuarial community, should be given very serious considerations when modelling dependent risks. In order to convey this message, we conduct a systematic study of the aforementioned construction. Specifically, we show, among other findings, that the model in Albrecher et al. (2011): (1) admits the interpretation of the Multiplicative Background Risk Model with risk components distributed gamma, and as such is easy to communicate to upper management; (2) is remarkably tractable, e.g., the risks aggregation within it is significantly simpler than in the case when the risk components are independent and distributed gamma; and (3) enjoys rich (tail) dependence characteristics.

Keywords: aggregation, background risk model, collective risk model, copulas, measure of dependence

JEL Classification: C02, C46

Suggested Citation

Semenikhine, Vadim and Furman, Edward and Su, Jianxi, On a Multiplicative Multivariate Gamma Distribution With Applications in Insurance (February 2, 2018). Available at SSRN: https://ssrn.com/abstract=3117055 or http://dx.doi.org/10.2139/ssrn.3117055

Vadim Semenikhine

York University ( email )

4700 Keele Street
Toronto, Ontario M3J 1P3
Canada

Edward Furman

York University - Department of Mathematics and Statistics ( email )

4700 Keele Street
Toronto, M3J 1P3
Canada

Jianxi Su (Contact Author)

Purdue University - Department of Statistics ( email )

West Lafayette, IN 47907
United States

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
69
Abstract Views
546
Rank
599,109
PlumX Metrics