Description: Automobile insurers classify the risks that they underwrite in order to subdivide the portfolio into classes of risks with similar profiles. While some insurers may have sufficient historical data, several others may not have significant volume of experience data in order to produce reliable claims predictions to help enhance their risk classification systems. A database containing a pooled experience of several insurers thereby helps to produce a more fair, reliable, and equitable premium structure for all risks concerned. This paper uses multilevel models to analyze the data on claim counts provided by the General Insurance Association of Singapore, an organization consisting of most of the general insurers in Singapore. The multilevel nature of the data is due to the following: a certain vehicle is observed over a period of years and is insured by a particular insurance company under a certain 'fleet' policy. The authors show how intercompany data lead to a priori premiums and a posteriori corrections to these initial premiums. Specific focus is made in understanding the intercompany effects using various count distribution models [Poisson, negative binomial, zero–inflated and hurdle Poisson]. The performance of these various models is compared the authors also investigate how to use the historical claims of a company, fleet and/or vehicle in order to correct for the premium initially set.

Description: This presentation develops a micro–level stochastic model for the run–off of automobile insurance claims. The analysis uses policy exposure, claims and payments experience data from vehicle insurance portfolios obtained from a major general insurance company in Singapore. The data are from the General Insurance Association 'GIA' of Singapore, an organization comprising most of the general insurers in Singapore. Modelling the run–off of individual Reported But Not Settled [RBNS] claims is the very aim of the work. The presenters focus on the following components of the development process: the time to an event, the conditional event type, the conditional payment type if any, and the conditional severity if any. They distinguish three types of events: settlement without any payments made during development, settlement with a payment and a payment without settlement. Given that a payment is to be made, the conditional payment type model describes the probability that it will be one of three claim types, or any possible combination of them. The conditional severity component describes the claim amount structure according to the combination of types paid. The presenters provide appropriate statistical models for each component and calibrate them to the available data. These building blocks are consequently used to predict the future development of open RBNS claims.