Towards a modular end-to-end statistical production process with mobile network data

Abstract

Mobile network data has proved to be an outstanding data source for the production of statistics in general, and for Official Statistics, in particular. Similarly to another new digital data sources, this poses the remarkable challenge of refurbishing a new statistical production process. In the context of the European Statistical System (ESS), we substantiate the so-called ESS Reference Methodological Framework for Mobile Network Data with a first modular and evolvable proposed statistical process comprising (i) the geolocation of mobile devices, (ii) the deduplication of mobile devices, (iii) the statistical filtering to identify the target population, (iv) the aggregation into territorial units, and (v) the inference to the target population. The proposal is illustrated with synthetic data generated from a network event data simulator developed for these purposes.