ORDERING PROPERTIES OF EXTREME CLAIM AMOUNTS FROM HETEROGENEOUS PORTFOLIOS
- 1 January 2019
- journal article
- research article
- Published by Cambridge University Press (CUP) in ASTIN Bulletin
- Vol. 49 (2) , 525-554
Abstract
In the context of insurance, the smallest and largest claim amounts turn out to be crucial to insurance analysis since they provide useful information for determining annual premium. In this paper, we establish sufficient conditions for comparing extreme claim amounts arising from two sets of heterogeneous insurance portfolios according to various stochastic orders. It is firstly shown that the weak supermajorization order between the transformed vectors of occurrence probabilities implies the usual stochastic ordering between the largest claim amounts when the claim severities are weakly stochastic arrangement increasing. Secondly, sufficient conditions are established for the right-spread ordering and the convex transform ordering of the smallest claim amounts arising from heterogeneous dependent insurance portfolios with possibly different number of claims. In the setting of independent multiple-outlier claims, we study the effects of heterogeneity among sample sizes on the stochastic properties of the largest and smallest claim amounts in the sense of the hazard rate ordering and the likelihood ratio ordering. Numerical examples are provided to highlight these theoretical results as well. Not only can our results be applied in the area of actuarial science, but also they can be used in other research fields including reliability engineering and auction theory.Keywords
This publication has 30 references indexed in Scilit:
- Comparisons of Parallel Systems According to the Convex Transform OrderJournal of Applied Probability, 2009
- Stochastic orders of scalar products with applicationsInsurance: Mathematics and Economics, 2008
- Heterogeneity and the need for capital in the individual modelScandinavian Actuarial Journal, 2006
- The total time on test transform and the excess wealth stochastic orders of distributionsAdvances in Applied Probability, 2002
- An Introduction to CopulasTechnometrics, 2000
- Convex orders for linear combinations of random variablesJournal of Statistical Planning and Inference, 2000
- The Dual Theory of Choice under RiskEconometrica, 1987
- Dispersive and superadditive orderingAdvances in Applied Probability, 1986
- Dispersive ordering and monotone failure rate distributionsAdvances in Applied Probability, 1985
- Inequalities: Theory of Majorization and its Applications.Journal of the American Statistical Association, 1981