Prognosegehalt von ifo-Geschäftserwartungen und ZEW-Konjunkturerwartungen: Ein ökonometrischer Vergleich
We compare the forecasting ability of the ifo-business expectations and ZEW-business expectations for the German industrial production in detail. Both are qualitative monthly surveys. While the ifo indicator is based on surveys of enterprises, the ZEW polls financial analysts from banks, insurances and large industrial companies. Using Granger causality tests we find a significant one-month lead of the ZEW-expectations over the ifo-expectations. Furthermore we find that the use of ZEW-expectations allows for longer term forecasts of German industrial produc tion. For a forecast period of three to twelve months the ZEW-expectations significantly outperform a naive forecast (which is simply based upon its own lagged values) of German industrial production as well as the forecasts based on the ifo-expectations. The ifo-expectations provide slightly better forecasts for short term periods (one month). We suggest that this difference in forecast ability is due to the different participants of the surveys: the financial analysts from the ZEW survey might incorporate more macroeconomic factors when building their expectations, thereby allowing longer term forecasts than the enterprises that take part in the ifo survey. However, the latter are supposed to forecast the immediate future better than the analysts as they have a better knowledge of the present situation of their business. Finally, using encompassing tests we show that a combination of both indicators results in better medium-term forecasts (three to six months) of the German industrial production than using both indicators alone.
Volume (Year): 222 (2002)
Issue (Month): 3 ()
|Contact details of provider:|| Postal: |
Phone: +49 (0)641 99 22 001
Fax: +49 (0)641 99 22 009
Web page: http://wiwi.uni-giessen.de/home/oekonometrie/Jahrbuecher/
More information through EDIRC
When requesting a correction, please mention this item's handle: RePEc:jns:jbstat:v:222:y:2002:i:3:p:316-336. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Peter Winker)
If references are entirely missing, you can add them using this form.