Robust estimation of revenues of Polish small companies by NACE section and province
Grażyna Dehnel , Łukasz Wawrowski
AbstractSample surveys conducted by the Central Statistical Office are currently the main source of information about revenues earned by small companies. Given the sample size, sampling scheme and the estimation method used in the survey, reliable estimates can only be produced for domains at the level of country, province or section of business classification. The market economy, however, creates a demand for local level information about businesses and economic conditions, which is provided on a regular basis at short intervals. The article describes an empirical study designed to test a small area estimation method. The goal of the study is to apply a robust version of the Fay-Herriot model, which, unlike the classical Fay-Herriot model, makes it possible to meet the assumption of normality of random effects under the presence of outliers. These alternative models will be supplied with auxiliary variables in order to estimate revenues of small businesses (with between 10 and 49 employees). Other sources of data used in the analysis include the DG1 report, Poland’s largest enterprise survey, and administrative registers. The study is expected to provide information about patterns and characteristics of the small business sector in Poland for territorial units on low level of aggregation.
|Publication size in sheets||0.5|
|Book||Papież Monika, Śmiech Sławomir (eds.): Proceedings of the 12th Professor Aleksander Zelias International Conference on Modelling and Forecasting of Socio-Economic Phenomena, Socio-Economic Modelling and Forecasting, no. 1, 2018, Foundation of the Cracow University of Economics, ISBN 978-83-65907-20-2, 611 p.|
|Keywords in English||small area estimation, indirect estimation, robust Fay - Herriot model, administrative registers, business statistics|
|Score||= 5.0, 12-03-2020, ChapterFromConference|
|Citation count*||2 (2020-09-24)|
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.