Neuer Artikel im Journal of Applied Statistics

Timo Schmid und Kollegen untersuchen die regionale Verteilung von Vermögen in Deutschland

The Fay-Herriot model for multiply imputed data with an application to regional wealth estimation in Germany

Kreutzmann, A.-K.; Marek, P.; Runge, M.; Salvati, N.; Schmid, T.

Abstract: The increasing inequality of private income and wealth requires the redistribution of financial resources. Thus, several financial support schemes allocate budget across countries or regions. This work shows how to estimate private wealth at low regional levels by means of a modified Fay-Herriot approach that deals with a) unit and item non-response, especially with used multiple imputation, b) the skewness of the wealth distribution, and c) inconsistencies of the regional estimates with the national direct estimate. One compelling example for financial redistribution is the promoted catching-up process of East Germany after the German reunification. This work shows that 25 years after the reunification differences are more diverse than just between the East and the West by estimating private wealth at two regional levels in Germany. The analysis is based on the Household Finance and Consumption Survey (HFCS) that the European Central Bank launched for all euro area countries in 2010.  Although the application in this paper focuses particularly on Germany, the approach proposed is applicable to the other countries participating in the HFCS as well as to other surveys that make use of multiple imputation.

Ann-Kristin Kreutzmann, Philipp Marek, Marina Runge, Nicola Salvati & Timo Schmid (2021): The Fay–Herriot model for multiply imputed data with an application to regional wealth estimation in Germany, Journal of Applied Statistics, DOI: 10.1080/02664763.2021.1941805