Membership / Notices to Members
Publications & Research
Research

New E-Forum Papers Published on Technology and Ratemaking

Last August, the CAS announced its 2025 Ratemaking Call Paper Program on Technology and the Ratemaking Actuary. The Ratemaking Working Group selected to support four papers from a wide array of impressive proposals, and those works are now published and available to the property-casualty actuarial community.

Head over to the CAS E-Forum to learn more about innovations, practical insights, and technological advances in ratemaking.

  • Winner of the 2025 Ratemaking Prize—“Enhancing Actuarial Ratemaking with Synthetic Data for Privacy Preservation,” by Noa Zamstein.
    Synthetic data preserves statistical integrity while protecting privacy, enabling accurate modeling and secure data sharing. The approach balances regulatory compliance with analytical utility, offering a promising path for privacy-preserving ratemaking.
  • Agent-Based Modeling: Introduction and Actuarial Applications,” by Rick Gorvett.
    Agent-based modeling is a valuable tool for capturing complex socioeconomic and risk processes, offering insights into macro-behaviors from micro-level actions and enhancing both risk quantification and strategy development.
  • Improving Trend Estimation Using Mix of Business Data,” by Mark Shapland and Trevor Parish.
    Changes in mix of business can impact observed trend data. Actuaries can adjust trend data for changes in the mix of business to more easily identify residual or “true” trends. This may lead to improved trend estimation and better understanding of a book of business.
  • Bridging Data Divides: AI as a New Paradigm for Unstructured Data,” by Sergey Filimonov.
    This paper presents a case study analyzing environmental reports from the U.S. Environmental Protection Agency to illustrate the practical application of AI techniques. By leveraging embeddings, structured information retrieval, and retrieval-augmented generation, actuaries can extract actionable insights from dense technical documents, demonstrating how AI can augment traditional methods in data-intensive scenarios.

Learn more about CAS research and additional research opportunities on the CAS Research Page.