Dealing with uncertainty in international migration predictions: from probabilistic forecasting to decision analysis

The paper focuses on the uncertainty of international migration predictions, as well as on their consequences for population projections and information delivered to decision makers. One of the key questions about probabilistic population forecasting is how its outcomes – the predictive distributions – can be useful for policy making and planning purposes. Some insights in that respect can be drawn from the statistical decision analysis, which takes into account the potential costs of both under- and overestimation of the variables under study, for example of current or future migration flows. The on-going paradigm shift in demographic projections, from deterministic to stochastic, can thus be brought even further, to the field of decision support. In that regard, the paper presents the preliminaries of Bayesian decision analysis together with some examples concerning international migration forecasts.