2020/02/02 by Rosy Oh, Oh, Rosy, Young-Ju Lee +5
Computer Science · Decision Sciences · Mathematics · #Applications (stat.AP) #Bayesian Methods and Mixture Models #FOS: Computer and information sciences #Probability and Risk Models #Statistical Methods and Inference
paper · pdf · doi:10.48550/arxiv.2002.00542
openalex publication_date 2020/02/02 · openalex created_date 2022/10/04 · openalex updated_date 2026/07/28
Typical risk classification procedure in insurance is consists of a priori\nrisk classification determined by observable risk characteristics, and a\nposteriori risk classification where the premium is adjusted to reflect the\npolicyholder's claim history. While using the full claim history data is\noptimal in a posteriori risk classification procedure, i.e. giving premium\nestimators with the minimal variances, some insurance sectors, however, only\nuse partial information of the claim history for determining the appropriate\npremium to charge. Classical examples include that auto insurances premium are\ndetermined by the claim frequency data and workers' compensation insurances are\nbased on the aggregate severity. The motivation for such practice is to have a\nsimplified and efficient posteriori risk classification procedure which is\ncustomized to the involved insurance policy. This paper compares the relative\nefficiency of the two simplified posteriori risk classifications, i.e. based on\nfrequency versus severity, and provides the mathematical framework to assist\npractitioners in choosing the most appropriate practice.\n