Every year, the IFoA recognises excellence in research and communicating actuarial knowledge by awarding prestigious prizes for outstanding papers published in IFoA publications. This year’s prizes, rewarding articles published in 2024, demonstrate excellence and impact across a range of areas encountered by actuaries in their practice and research.
The Geoffrey Heywood Prize (named for a former President of the Institute of Actuaries who established the prize with a bequest in his will) is awarded for excellent levels of communication and engagement with a general actuarial audience. The recipient this year is Caesar Balona for his British Actuarial Journal (BAJ) paper ‘ActuaryGPT: applications of large language models to insurance and actuarial work’. The paper has been the subject of an IFoA sessional meeting.
The judging panel agreed this highly topical paper provided an excellent entry point into the world of large language models (LLMs) for actuaries wishing to learn more about this technology and the potential it holds for actuarial applications. It engages directly with the demand for a deeper understanding of how LLMs can function in insurance settings.
On receiving the prize, Caesar said: “When ChatGPT launched in late 2022, I recognised that the underlying technology would fundamentally transform how humans work. I searched for resources exploring its potential impact on the actuarial profession and how we might harness these tools but found little available guidance. I decided to fill this gap by learning about LLMs myself and creating a resource I felt our profession needed. Receiving the Geoffrey Heywood Prize validates that I achieved this goal. It is both a profound honour and tremendous encouragement for my work.”
Highly commended awards were given to:
Both were published in the BAJ.
The Peter Clark Prize (in memory of a former President of the Institute of Actuaries) is awarded for an outstanding paper published for an actuarial audience. This year’s prize winners are Nhan Huynh and Mike Ludkovski for their Annals of Actuarial Science (AAS) paper ‘Joint models for cause-of-death mortality in multiple populations’.
The panel noted this topic is both technically challenging and of great current importance. The authors address the challenge using cutting-edge spatial machine learning methods, with the paper combining an impactful application with methodological rigour and innovation.
“We are honoured to be recognised by the IFoA for our work on cause-of-death mortality in a multinational setting”, Mike commented. “Our paper leverages the Human Cause-of-Death Database to develop joint stochastic models that efficiently capture cross-cause and cross-population dependencies. Such information fusion boosts predictive power to smooth and extrapolate cause- and age-specific mortality rates, providing insights into commonalities and heterogeneities in mortality trends across causes, countries and genders.”
A highly commended award was given to Ronald Richman and Mario V Wüthrich for their AAS paper ‘Smoothness and monotonicity constraints for neural networks using ICEnet’.
If you’re inspired to write for potential publication in the IFoA’s journals, visit the:
These pages give more information about their scope and how to submit your work.
