Guy J. Abel , University of Hong Kong
Peter W. F. Smith, University of Southampton
Bilateral migration data, summarizing the number of people migrating between each origin and destination, provide a clearer understanding of migration patterns than summary measures such as net migration. However, bilateral migration flow data are commonly unavailable, not up to date or provide conflicting accounts of population change when compared to changes implied in demographic data. In this paper we introduce an conditional maximisation routine to update bilateral migration data to match known net migration totals. The routine can also be applied in missing data situations to generate synthetic bilateral migration flows to match known net migration totals where no reported bilateral data are available. We illustrate the method using real world data from the United States. The resulting estimates of bilateral migration flows are demographically consistent with changes in reported population totals, births and deaths over the period and provide a detailed depiction of contemporary state to state migration patterns.
Presented in Session 29. Estimation Methods