Probabilistic Method for Combining Internal Migration Data
Guy Abel1, Guillermo Vinue2, Dilek Yildiz2, Arkadiusz Wisniowski3
1Asian Demographic Research Institute, Shanghai University, Shanghai, China. 2Wittgenstein Centre for Demography and Global Human Capital (International Institute for Applied Systems Analysis, Vienna Institute of Demography/Austrian Academy of Sciences, Vienna University of Economics and Business), Vienna, Austria. 3Cathie Marsh Institute for Social Research, School of Social Sciences, University of Manchester, Manchester, United Kingdom

Abstract
In order to fully understand the causes and consequences of population movements, researchers and policy makers require timely and consistent data. Migration data are commonly obtained from censuses, registers or surveys. Each of these data sources can vary in their measurement of accuracy, coverage of population, undercount and definitions of a migration event. This paper proposes a Bayesian probabilistic methodology to harmonize migration data from different sources. In particular, we build a hierarchical model for combining migration data sources in the USA between 1980 and 2016. The model allows for estimates of true migration flows that explicitly compensates for the inadequacies in each data source and provides one-step ahead forecasts of bilateral migration patterns.