CAS Releases Two Requests for Proposals on Emerging Ratemaking Research Topics

The CAS and the CAS Ratemaking Working Group have released two new requests for proposals on emerging ratemaking topics: scaling laws in pure premium models and severe convective storm event exposures. The detailed topics are as follows:

Scaling Laws in Premium Models

This request looks to address the debate about how to best improve large language pure premium models moving forward. A key piece of research that led to a push for ever-increasing datasets and network sizes was Scaling Laws for Neural Language Models (2020, Kaplan et al). The call proposes that actuaries carry out similar research for pure premium modeling in an insurance context. The resulting scaling law would serve as a benchmark for all interested parties on model performance across a range of dataset sizes.   

Proposals on Severe Convective Storm (“SCS”) Exposures

This call explores methods and techniques that incorporate exposures related to SCS events in the ratemaking process. There are a wide variety of papers available for events such as hurricanes, wildfires, and earthquakes. However, ratemaking for SCS can be challenging, as these events can cause damage to insured properties stemming from multiple natural elements including lightning, tornadoes, hail, and straight-line wind, all of which are difficult to predict from a ratemaking perspective. The group is, therefore, seeking to broaden the available material on the topic of pricing for SCS exposures addressing either a primary carrier perspective or reinsurance implications.

The deadline for proposals for both projects is Wednesday, January 31, 2024. Please submit any questions to Elizabeth Smith, CAS director of publications and research, at esmith@casact.org.