A Bayesian Spatial Modeling Approach to Mortality Forecasting

TL;DR

This paper introduces a Bayesian spatial model for mortality forecasting that accounts for geographic autocorrelation and overdispersion, demonstrating improved predictive accuracy across Japanese counties.

Contribution

It extends Bayesian mortality models by incorporating spatial autocorrelation and overdispersion, providing a more flexible and accurate forecasting framework.

Findings

The model effectively captures spatial autocorrelation in mortality data.

It outperforms traditional models in predictive accuracy.

The approach is demonstrated on Japanese county data across multiple years.

Abstract

This paper extends Bayesian mortality projection models for multiple populations considering the stochastic structure and the effect of spatial autocorrelation among the observations. We explain high levels of overdispersion according to adjacent locations based on the conditional autoregressive model. In an empirical study, we compare different hierarchical projection models for the analysis of geographical diversity in mortality between the Japanese counties in multiple years, according to age. By a Markov chain Monte Carlo (MCMC) computation, results have demonstrated the flexibility and predictive performance of our proposed model.