The MIMOSA model for estimating international migration flows in the European Union

James Raymer, University of Southampton

MIMOSA (MIgration MOdelling for Statistical Analyses) is a three-year project funded by Eurostat to support the development and application of statistical modelling techniques for the estimation of missing data on migration flows and foreign population stocks. This paper presents the modelling approach adopted by the MIMOSA project to estimate international migration flows between countries in the European Union. The aim is to account for the many differences in definitions, quality and sources of available migration data and to estimate the missing data. The methodology adopted includes two stages. The first uses a simple mathematical procedure to harmonise the available data. The second estimates the missing marginal data and associations between countries by using the available (harmonised) flows and covariate information. Both stages are set within a multiplicative framework for analysing migration flows.