A meta-analytic framework for best-practice mortality surface estimation
Magdalena Muszynska, Duke University
Jutta Gampe, Max Planck Institute for Demographic Research
Paul H.C. Eilers, Utrecht University
Similar to the best-practice life expectancy, the best-practice mortality surface over age and time can be estimated. Reliability is a serious issue, because population size across countries varies over several orders of magnitude. Our solution is to estimate a latent mortality distribution, properly accounting for observed magnitudes. This brings a meta-analytic framework to mortality studies, opening many new opportunities to study variability between demographic units in a quantitative way.
Presented in Session 50: Methodological Issues in Mortality