The aim of this study was to develop a methodological tool to ‘nowcast’ migrant stocks by using real-time
data from the Facebook Marketing Application Programming Interface (API) and official migration data
from EU member states and states in the United States. To achieve this, two specific objectives were set:
(i) To collect real-time data that could provide estimates of migrant stocks in the countries of
interest from the Facebook Marketing API, along with migrant-stocks data from official
sources from 2010 onwards (including meta information about the quality of these data).
(ii) To develop a Bayesian model capable of combining real-time data from the Facebook
Marketing API and official migration data to nowcast stocks of migrants in EU member
states and US states, and test the model in some corridors, using real-world data. This
approach built on similar research undertaken by RAND to develop a proof-of-concept
for nowcasting labour mobility within the EU.
Central to the methodological approach in this study was the availability of real-time proxy estimates of international migrant stocks using data from the Facebook Marketing API, specifically the ‘Lived in’ status, which is used as a proxy for country of origin. In total, 22 months of this data was collected from the Facebook Marketing API (from December 2019 to September 2021). In addition, a range of official data containing estimates of international migrant stocks from various data sources was collected, including data from Eurostat, the OECD, the UN and the US Census Bureau. A detailed metadata review of each dataset was carried out to understand the key characteristics of the data and assess the main strengths and limitations.

Sodipress vous remercie de renseigner votre email pour recevoir le document.
Envoyer le document
You May Also Like