demofit: A Unified Framework for Mortality Modelling Integrating Parametric Laws and Stochastic Models
Year: 2026
Type: article
Abstract: Mortality modelling and forecasting are widely used in actuarial science and demography, with numerous parametric mortality laws and stochastic mortality models proposed to describe agespecific patterns and temporal dynamics.Existing software implementations, however, typically focus on particular model classes and do not provide a unified framework for combining these approaches.We present the R package demofit, which integrates parametric mortality laws and stochastic mortality models within a coherent and user-friendly interface.The package includes a broad collection of mortality laws, together with widely used forecasting models such as the Lee-Carter, Cairns-Blake-Dowd, and spatial-temporal vector autoregressive frameworks.Model estimation is performed using specific optimisation routines to enhance robustness, and forecasts can be further regularised using parametric mortality curves to produce smooth and interpretable age profiles.We describe the underlying framework and illustrate the main functions of the package using real mortality data, demonstrating its application in model fitting, forecasting, and analysis.
Source: The R Journal
DOI: 10.32614/rj-2026-053
Author: Jackie Li
Institution: Singapore Management University
Language: English
Subfield: Demography
Field: Social Sciences
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W7214926321
