Abstract
The compound binomial model is a discrete time analogue (or approximation) of the compound Poisson model of classical risk theory. In this paper, several results are derived for the probability of ruin as well as for the joint distribution of the surpluses immediately before and at ruin. The starting point of the probabilistic arguments are two series of random variables with a surprisingly simple expectation (Theorem 1) and a more classical result of the theory of random walks (Theorem 2) that is best proved by a martingale argument.
Keywords Ruin; binomial model; compound binomial process; bonus malus; severity of ruin.
Volume
18:2
Page
161-168
Year
1988
Categories
Actuarial Applications and Methodologies
Ratemaking
Classification Plans
Financial and Statistical Methods
Risk Pricing and Risk Evaluation Models
Probability of Ruin
Financial and Statistical Methods
Loss Distributions
Publications
ASTIN Bulletin