|
Abstract:
|
This work describes statistical modeling of detailed, micro-level auto insurance records from a major insurer. By micro-level records, we mean claims experience at the individual vehicle level with detailed information on the type of claim, such as whether it is due to injury to a third party, property damage to a third party or claims for damage to the insured, as well as the corresponding claim amount. We propose a hierarchical model for three components: the frequency, type and severity of claims. We show a significant dependence among the claim types, using a t-copula to account for dependence. The model justifies assessing the importance of a rating variable allowing for more efficient prediction of claims compared to traditional methods. Using simulation, we demonstrate this by developing predictive distributions and calculating premiums under alternative coverage limitations.
|
