Bayesian Modeling of Demographic Data in Many Subpopulations

Analyzing age-specific mortality, fertility, and migration in subpopulations is a crucial task in demography, with significant policy relevance. In practice, such analysis is challenging when studying numerous subpopulations, due to small sample sizes and demographic heterogeneity. To address this issue, a Bayesian model for the joint analysis of many, potentially small, demographic subgroups is proposed. The model combines three common assumptions about demographic processes in a unified probabilistic framework. The approach provides robust estimates of the demographic process in each subpopulation, allows testing for heterogeneity between subpopulations, and can be used to assess the impact of covariates on the demographic process. This makes the model suitable for probabilistic projection exercises and scenario analysis. An in-depth analysis of age-specific immigration flows to Austria, disaggregated by sex and 155 countries of origin, is used to illustrate the framework. Comparative analysis shows that the model outperforms commonly used benchmark frameworks in both in-sample imputation and out-of-sample prediction exercises.