Bayesian Forecasting of Immigration to Selected European Countries by using Expert Knowledge

Summary 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 barely predictable in the long run, exhibiting non-stationary features. This outcome was obtained largely irrespectively 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.