Research Programme


Evaluation of existing migration forecasting methods and models

Jakub Bijak, George Disney, Allan FindlayJon ForsterPeter W.F. Smith,  Arek Wiƛniowski

Project summary

Research commissioned by the Migration Advisory Committee, Home Office

This project aims to evaluate the various existing approaches to modelling and forecasting of UK international migration flows, based on the literature, overview of official forecasts prepared in other countries, and own empirical analysis. The empirical assessment is done by comparing the results of various forecasting models for selected migration flows against the trends observed in the past. The overarching goal is to assess the uncertainty of different migration forecasting methods.

The research will contribute to the evidence base in migration research by providing a systematic and up to date methodological overview of academic literature and of official statistical practice of migration modelling and forecasting, strengthened by the empirical comparison of performance of various methods. The work will also provide concrete policy recommendations related to the usefulness of various forecasting approaches for policy-makers, with focus on the role of uncertainty and its appropriate communication to the forecast users.

Project activities
 

Date Activity Description
15 September 2015 Presentation on “Methods for Forecasting Migration: Evaluation and Policy Implications”, Migration Statistics User Forum conference  Summary talk with the final project results, given by Jakub Bijak
24 April 2015 Preliminary results Presentation to the Migration Advisory Committee by George Disney and Jakub Bijak

 


 

 

 

 

Publications

Please check back soon for publications from this project. Alternatively, browse all CPC publications.

Media activities

Migration is one of the worst predictable elements of the puzzle - Population Europe Inter-Faces video interview with Jakub Bijak.

You can also browse population-related articles from CPC members on our Scoop.it! page.