This Detranote introduces a novel parametric framework for modeling regional mortality differences across European NUTS-2. Because regional death rates are substantially noisier than national aggregates, we extend the Li and Lee model by considering penalized B-splines for smoothing central log-death rates and marginal age-specific effects of the endogenous time factors.
Compared to the Li and Lee framework, our model offers several advantages. First, it yields smooth mortality curves of mortality rates, which is a feature highly desirable for actuarial applications. Second, the parameterization limits the model’s degrees of freedom, enabling reliable estimation over shorter time windows. Thirdly, the model is sufficiently robust for long-term mortality forecasting and annuity valuation.
Model parameters are estimated via maximization of the Poisson log-likelihood, and the methodology is applied to French and Belgian regional mortality data. A companion Python package has been developed for the purpose of this analysis and is freely available on GitHub.
Keywords: mortality model, Lee-Carter model, Li-Lee model, multi-group mortality, life insurance.
Sector: Insurance
Domain: Life
Authors: I. Belem, M. Denuit & T. Hames
Publisher: Detralytics
Date: April 2026
Language: English
Pages: 26
Reference : Detra Note 2026-2
Idrissa is an Actuarial Consultant and part of the Talent Accelerator Program (TAP) at Detralytics. With two years of work-study experience, Idrissa has developed solid expertise in non-life insurance, particularly in pricing and CAT modeling. His diverse experiences have enabled him to gain strong proficiency in several tools and further strengthen his technical expertise.
Donatien Hainaut is a Scientific Advisor at Detralytics and a professor at UCLouvain (Belgium), where he serves as the Director of the Master’s program in Data Science with a statistical orientation. Prior to this, he held several academic positions, including Associate Professor at Rennes School of Business and ENSAE in Paris. He also has extensive industry experience, having worked as a Risk Officer, Quantitative Analyst, and ALM Officer.
Donatien is a Qualified Actuary and holds a PhD in the field of Asset and Liability Management. His current research focuses on contagion mechanisms in stochastic processes and the applications of neural networks in insurance.
Since joining Detralytics in 2019, Thomas has primarily worked on pricing and reserving topics in P&C retail Auto and Non-Life, combining classical methods with ML-based approaches.
Additionally, internally, Thomas is involved in the technical and commercial development of our micro-reserving package and ensures the monitoring and supervision of projects emerging within our Innovation Lab.
