Multivariate risk model of phase type [An article from: Insurance Mathematics and Economics]
Book Details
Author(s)J. Cai, H. Li
PublisherElsevier
ISBN / ASINB000RR1TEW
ISBN-13978B000RR1TE9
AvailabilityAvailable for download now
MarketplaceUnited States 🇺🇸
Description
This digital document is a journal article from Insurance Mathematics and Economics, published by Elsevier in 2005. The article is delivered in HTML format and is available in your Amazon.com Media Library immediately after purchase. You can view it with any web browser.
Description:
This paper is concerned with several types of ruin probabilities for a multivariate compound Poisson risk model, where the claim size vector follows a multivariate phase type distribution. First, an explicit representation for the convolution of a multivariate phase type distribution is derived, and then an explicit formula for the ruin probability that the total claim surplus exceeds the total initial reserve in infinite horizon is obtained. Furthermore, the effect of the dependence among various types of claims on this type of ruin probability is considered under the convex and supermodular orders. In addition, the bounds for other types of ruin probabilities are developed by utilizing the association of multivariate phase type distributions. Finally, some examples are presented to illustrate the results.
Description:
This paper is concerned with several types of ruin probabilities for a multivariate compound Poisson risk model, where the claim size vector follows a multivariate phase type distribution. First, an explicit representation for the convolution of a multivariate phase type distribution is derived, and then an explicit formula for the ruin probability that the total claim surplus exceeds the total initial reserve in infinite horizon is obtained. Furthermore, the effect of the dependence among various types of claims on this type of ruin probability is considered under the convex and supermodular orders. In addition, the bounds for other types of ruin probabilities are developed by utilizing the association of multivariate phase type distributions. Finally, some examples are presented to illustrate the results.
