Bayesian Cohort Component Population Projections

Peter W.F. Smith, University of Southampton
Arkadiusz Wisniowski, University of Southampton
James Raymer, Australian National University

In this paper, we explore the use of Bayesian methods for projecting the United Kingdom's age- and sex-specific population. We first argue that a Bayesian approach is a more natural framework for incorporating various forms of uncertainty in probabilistic projections. Second, we demonstrate the consequences of choosing different Lee-Carter type models for fertility, mortality, immigration and emigration in terms of forecasted age patterns and their associated measures of uncertainty. Third, we incorporate these forecasts into a cohort component projection model and compare the results. We end the paper by discussing the merits and flexibility of a Bayesian cohort component projection model and highlight some areas where this work could be extended.