Small area estimation for city statistics and other functional geographies

The city data collection is one of the regular data collections of Eurostat and the national statistical institutes. The demand for timely and reliable socio-economic data on cities and functional urban areas has significantly increased. Since 2017, the cities and their functional urban areas are legally recognised by the amended NUTS regulation. To produce socio-economic data coming originally from sample surveys at the level of small units such as cities, functional urban areas and other functional geographies is a complex task which requires the application of small area estimation techniques since those functional geographies are usually not incorporated in the sampling design. This paper aims at investigating on the performance of different estimation strategies against the background of various sampling designs used by the national statistical institute. Therefore, a design-based Monte Carlo simulation study using a synthetic but close-to-reality population has been performed. The results show that all investigated estimation approaches are able to increase the efficiency and the quality of the estimates compared to the classical design-weighted estimation techniques.