Online Program

Return to main conference page

All Times ET

Program is Subject to Change

Tuesday, June 15
Tue, Jun 15, 11:30 AM - 1:00 PM
Topics in the Collection, Production, and Estimation of Short Term and other Business Statistics

Small Area Estimation for the Job Statistics, an example of innovative experimental statistics from SFSO (308184)

Monika Ferster, Swiss Federal Statistical Office 
*Daniel Kilchmann, Swiss Federal Statistical Office 
Anne Massiani, Swiss Federal Statistical Office 
Djalel-Eddine Meskaldji, Swiss Federal Statistical Office 

Keywords: Small area estimation, design-based validation, job statistics, experimental statistics, data innovation

Similar to other National Statistical Institutes and Eurostat, the Swiss Federal Statistical Office (SFSO) decided to promote data innovation through a website dedicated to experimental statistics,, exploring the prospects to enlarge and complement official statistics. One of the pilot projects presented on this website and emerging from SFSO’s data innovation strategy evaluates the potential of the use of Small Area Estimation (SAE) methods for the Job Statistics. The aim of this pilot project is to produce quarterly statistics of the Job Statistic on detailed levels such as small regions (cantons) and NACE, which could not be published using direct estimation due to unsatisfactory precision. The use of the weights of the establishment units and the development of a design-based validation procedure were the main methodological challenges considered in this pilot project. Moreover, bootstrap techniques had been applied to obtain an approximation of the design-based variance. Furthermore, the validation had to take into account the stability of the time series covering the twelve quarters of the years 2015 to 2017. This contribution outlines the main findings tackling these methodological challenges and shows the resulting outcomes. A concluding discussion with respect to the pertinence and the benefits of innovative experimental statistics such as the use of SAE methods for establishment surveys is closing this contribution.