The Simulation Method Of Reserve Risk Based On RJMCMC

The process of setting claim reserves for the outstanding loss liabilities is an important part of the management of general insurance companies. With the increasing uncertainty,only a single provisioning results do not meet the requirements of the regulatory authorities and corporate risk management. A method is needed to evaluate the risk of reserve. On the one hand the traditional stochastic models can provide reserve estimates, and on the other hand it can provide a measure of variability in the reserve estimates.Although it is hard to obtain the predictive distributions of outstanding liabilities in general insurance analytically, it can be obtained using bootstrap or Bayesian techniques for clearly defined statistical models.With few exceptions, the literature on non-life insurance reserve risk has been devoted to the ultimo risk, the risk in the full run-of the liabilities. This is in contrast to the short time horizon in models for the total risk of the insurance company, and in particular the one-year risk perspective taken in the Soiency Ⅱ project This paper aims at clarifying the methodology for the one-year risk; in particular we describe a simulation approach to the one-year reserve risk.We introduce the Bayesian overdispersed Poisson (ODP) model in order to estimate the reserve and we describe in carefull detail the RJMCMC method and its application to Bayesian ODP model.Then, we use claim development result(CDR) and re-reserve(RR) based on re-reserving to describe one-year reserve risk whick can be beneficial to management jugement.