Common mortality modeling and coherent forecasts. An empirical analysis of worldwide mortality data

A new common mortality modeling structure is presented for analyzing mortality dynamics for a pool of countries, under the framework of generalized linear models (GLM). The countries are first classified by fuzzy c-means cluster analysis in order to construct the common sparse age-period model structure for the mortality experience. Next, we propose a method to create the common sex difference age-period model structure and then use this to produce the residual age-periodmodel structure for each country and sex. The time related principal components are extrapolated using dynamic linear regression (DLR) models and coherent mortality forecasts are investigated. We make use of mortality data from the “Human Mortality Database”.

► We classify 35 countries (utilizing HMD) by fuzzy c-means cluster analysis. ► We pool 19 countries to construct the post-war common sparse age-period structure. ► We derive two basic sparse terms; the main trend and the young-adults relative trend. ► We model the common sex differentials, giving a coherent structure between them. ► The two common sparse terms and the sex differentials offer coherent forecasts.