Measuring sensitivity in a bonus-malus system

Abstract

In performing Bayesian analysis of a bonus-malus system (BMS) it is normal to choose a parametric structure, 70(.), in the insurer's portfolio. According to Bayesian sensitivity analysis the structure function can be modelled by specifying a class F of priors instead of a single prior. In this paper, we examine the ranges of the relativities, i.e. delta(pi) = E[lambdapi (lambda\data)/E[lambdapi(lambda)], pi is an element of Gamma. We illustrate our method with data from [Astin Bulletin 10 (3) (1979) 274]. (C) 2002 Elsevier Science B.V. All rights reserved.