Bayesian demography 250 years after Bayes

Abstract

Bayesian statistics offers an alternative to classical (frequentist) statistics. It is distinguished by its use of probability distributions to describe uncertain quantities, which leads to elegant solutions to many difficult statistical problems. Although Bayesian demography, like Bayesian statistics more generally, is around 250 years old, only recently has it begun to flourish. The aim of this paper is to review the achievements of Bayesian demography, address some misconceptions, and make the case for wider use of Bayesian methods in population studies. We focus on three applications: demographic forecasts, limited data, and highly structured or complex models. The key advantages of Bayesian methods are the ability to integrate information from multiple sources and to describe uncertainty coherently. Bayesian methods also allow for including additional (prior) information next to the data sample. As such, Bayesian approaches are complementary to many traditional methods, which can be productively re-expressed in Bayesian terms.

Childlessness and fertility by couples' educational gender (in)equality in Austria, Bulgaria, and France

“…Bijak and Bryant (2016) recently carried out an overview of the use of Bayesian methods in demography.…”

Spatial demography: A unifying core and agenda for further research

“…Raymer, Abel and Rogers (2012) utilised vector autoregressive models to capture the serial autocorrelation and spatial dependency in demographic components across regions in England, but their models did not account for age or sex. Finally, the recent developments in Bayesian methods applied to demographic estimation and projection have much potential for both transparency and ability to both borrow strength across patterns in the data (Bijak & Bryant, 2016;Raftery, Li, Ševčíková, Gerland, & Heilig, 2012).…”

Global estimation of neonatal mortality using a Bayesian hierarchical splines regression model

“…Increases in computational speed as well as the development of suitable numerical methods has enabled a more widespread use of a Bayesian approach in many fields, including population estimation and forecasting (e.g., Alkema and New 2014;Bijak and Bryant 2016;Girosi and King 2008;Raftery et al 2012;Schmertmann et al 2014). The method presented in this paper has similarities to approaches used to estimate other global health indicators, including the U5MR (Alkema and New 2014;You et al 2015), maternal mortality (Alkema et al 2016), cause-specific mortality (Foreman et al 2012), and contraceptive prevalence (Alkema et al 2013).…”