Abstract
In this paper we mirror the framework of generalized (non-)linear models to define the family of generalized age-period-cohort stochastic mortality models which encompasses the vast majority of stochastic mortality projection models proposed to date, including the well-known Lee-Carter and Cairns-Blake-Dowd models. We also introduce the R package StMoMo which exploits the unifying framework of the generalized age-period-cohort family to provide tools for fitting stochastic mortality models, assessing their goodnessof-fit and performing mortality projections. We illustrate some of the capabilities of the package by performing a comparison of several stochastic mortality models applied to the England and Wales population.
Highlights
During the last two centuries developed countries experienced a persistent increase in life expectancy
We introduce the R package StMoMo (Villegas, Millossovich, and Kaishev 2018) which exploits the unifying framework of the generalized ageperiod-cohort family combined with the powerful fitting function of the gnm package (Turner and Firth 2015) to provide computational tools for implementing many of the stochastic mortality models proposed to date
Akin to generalized linear models, a generalized age-period-cohort (GAPC) stochastic mortality model is comprised of four components: 1. The random component: The numbers of deaths Dxt follow a Poisson distribution or a Binomial distribution, so that
Summary
During the last two centuries developed countries experienced a persistent increase in life expectancy. Package StMoMo comes with a set of functions for defining an abstract model – specifying for instance the number of period terms, whether coefficients are parametric or not – and for fitting a given model. This is useful when estimating several models on a given dataset or a given model to different datasets. In this paper we describe the statistical framework underlying StMoMo and illustrate its usage For this purpose, we use as a running example a comparison of several stochastic mortality models fitted to the England and Wales population.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have