2021 Hachemeister Prize Announcement
The 2021 Hachemeister Prize has been awarded to the paper “AGLM: A Hybrid Modeling Method of GLM and Data Science Techniques” by Suguru Fujita, Toyoto Tanaka, Kenji Kondo, and Hirokazu Iwasawa. The Hachemeister Prize is awarded annually to ASTIN Bulletin or ASTIN or AFIR Colloquium paper(s) based on several criteria but with emphasis placed on the paper's impact for North American actuaries and its practicality of application.
The committee believed “AGLM: A Hybrid Modeling Method of GLM and Data Science Techniques” was impactful and valuable for Actuaries to read because:
- the paper’s hybrid modeling approach addresses a real need to balance the accuracy of data science techniques with the strong explanatory power of GLMs
- the authors evaluate the advantages of the AGLM method both qualitatively and quantitatively against existing modeling methods, using an example in automobile insurance
- the paper is accompanied by a package in the statistical software R to allow practicing actuaries to use the AGLM method easily, and
- while the paper presents an example in pricing automobile insurance, the concepts presented may extend to other property/casualty lines or other applications.
Overall, the committee believed this paper would be interesting to all actuaries, includes practical uses and original thought, and is well written.
The authors have been invited to present their prize-winning paper at the 2021 CAS Annual Meeting November 7-10, 2021.
There were many good, eligible papers this year, and the committee wanted to specifically recommend the following paper for additional reading for actuaries involved in pricing, but may be of general interest:
- “Discrimination-Free Insurance Pricing” by Mathias Lindholm, Ronald Richman, Andreas Tsanakas, and Mario Wüthrich is a good paper that proposes a procedure for incorporating certain individual (discriminatory) policyholder information to prevent both direct and indirect discrimination.
Furthermore, the committee thought the following papers would make for valuable additional reading for actuaries interested in the emerging cyber field:
- “Silent cyber assessment framework” by the Institute and Faculty of Actuaries’ Cyber Risk Investigation Working Party (Simon Cartagena, Visesh Gosrani, Jasvir Grewal, and Justyna Pikinska) is a good paper that proposes a framework to help insurance companies address the issue of non-affirmative cyber risk across their portfolios.
- “Cyber claim analysis through Generalized Pareto Regression Trees with applications to insurance” by Sébastien Farkas, Olivier Lopez, and Maud Thomas is a good paper that proposes a method based on regression trees to analyze cyber claims to identify criteria for claim classification and evaluation.