Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/73697 
Authors: 
Year of Publication: 
2007
Series/Report no.: 
ifo Working Paper No. 40
Publisher: 
ifo Institute - Leibniz Institute for Economic Research at the University of Munich, Munich
Abstract: 
Results from business tendency surveys are often used to construct leading indicators. The indicators are then, for example, employed to forecast GDP growth. In this article more detailed results of business tendency surveys are used to forecast quarter-onquarter GDP growth. The target series is very challenging because this type of growth rate leads to quite volatile time series. The present study focuses on German GDP data and survey results provided by the Ifo Institute. Since numerous time series of possible indicators result from the surveys, methods that can handle this setting are applied. One candidate method is principal component analysis, which is used to reduce dimensionality. On the other hand, subset selection procedures are applied. For the present setting the latter method seems more successful than principal components. But this is not a statement about the two types of procedures in general. Which method should be favoured depends very much on the aims of the specific study.
Subjects: 
Business tendency surveys
business cycle analysis
principal component regression
subset selection
JEL: 
C22
C42
E32
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.