Alternative Estimation Framework Based on GLM for Mortality Models: From Single Population to Cause-of-Death

Antoine Burg and Christophe Dutang ()
Additional contact information
Antoine Burg: CEREMADE - CEntre de REcherches en MAthématiques de la DEcision - Université Paris Dauphine-PSL - PSL - Université Paris Sciences et Lettres - CNRS - Centre National de la Recherche Scientifique
Christophe Dutang: ASAR - Applied Statistics And Reliability - ASAR - LJK - Laboratoire Jean Kuntzmann - Inria - Institut National de Recherche en Informatique et en Automatique - CNRS - Centre National de la Recherche Scientifique - UGA - Université Grenoble Alpes - Grenoble INP - Institut polytechnique de Grenoble - Grenoble Institute of Technology - UGA - Université Grenoble Alpes

Working Papers from HAL

Abstract: For actuarial purposes, commonly used mortality models break down the all-cause mortality into several components of age and time. Standard practice assumes a count distribution and uses Maximum Likelihood Estimation (MLE) for parameters estimation. Such models can be extended in various ways to model multi-populations, and similar ideas are also suitable to model cause-of-death (CoD) mortality. In some cases, the modelling structure allows an interpretation as a Generalized Linear Model (GLM). We leverage a GLM-formulation based on categorical explanatory variables to derive closed-form formulas for model parameters. This alternative estimator can be applied to classical all-cause mortality models with linear structure, and may in some cases even be preferred to MLE estimators. Above all, this estimation procedure can be generalized to handle multi-populations model and multivariate distributions for CoD mortality.

Date: 2024-10-30
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hal:wpaper:hal-04959839

DOI: 10.2139/ssrn.4990943

Access Statistics for this paper

More papers in Working Papers from HAL
Bibliographic data for series maintained by CCSD ().