Bayesian forecasting of immigration to selected European countries using expert knowledge

Jakub Bijak, University of Southampton
Arkadiusz Wisniowski, University of Southampton

The aim of the paper is to present Bayesian forecasts of immigration for seven European countries to 2025, based on quantitative data and qualitative knowledge elicited from country-specific migration experts in a two-round Delphi survey. In line with earlier results, most of the immigration processes under study were found to be hardly predictable in the long run, exhibiting non-stationary features. This outcome was obtained largely irrespective of the expert knowledge input, which nevertheless was found useful in describing the predictive uncertainty, especially in the short term. It is argued that under the non-stationarity of migration processes, too long forecasts horizons are inadequate, which is a serious challenge for population forecasts in general.

Presented in Session 40: Modelling demographic change in the context of uncertainty