Forecasting Economic Activity in Germany - How Useful are Sentiment Indicators?

Abstract

Die Discussion Papers dienen einer möglichst schnellen Verbreitung von neueren Forschungsarbeiten des ZEW. Die Beiträge liegen in alleiniger Verantwortung der Autoren und stellen nicht notwendigerweise die Meinung des ZEW dar.Discussion Papers are intended to make results of ZEW research promptly available to other economists in order to encourage discussion and suggestions for revisions. The authors are solely responsible for the contents which do not necessarily represent the opinion of the ZEW.Download this ZEW Discussion Paper from our ftp server:ftp://ftp.zew.de/pub/zew-docs/dp/dp0256.pdf Non-technical SummaryWe analyze four economic sentiment indicators for the German economy: the ifo business expectations (ifo), the European Commission's Economic Sentiment Ind icator for Germany (ESIN), the Purchasing Managers' Index (PMI) and the ZEW Indicator of Economic Sentiment (ZEW) using the year-on-year growth rate of i ndustrial production as a reference. A look at the publication schedule of the indic ators shows that all of them are released at least one month prior to the German industrial production statistics. Cross correlations indicate that the correlation between the ifo, PMI and ZEW indicators and the year-on-year growth rates of industrial production increases when the indicators are lagged. This suggests that these three indicators are indeed leading economic activity in Germany. The ESIN indicator, on the other hand, seems to lag industrial production.Subsequent Granger causality tests reveal that the ifo business expectations, the PMI and the ZEW Indicator of Economic Sentiment lead the year-on-year growth rate of German industrial production by five months. Taking into account the publication lag of industrial production (about six weeks) this lead extends to more than six months. Analyzing lead/lag structures among the ifo, PMI and ZEW indicators we find that the ZEW indicator significantly leads the ifo business expectations by one month. Furthermore, the ifo expectations indicator has a lead of one month over the PMI.Out-of-sample forecast evaluations suggest that both ifo and ZEW provide the best forecasts for industrial production among the three indicators ifo, PMI and ZEW. While the ZEW indicator performs better than the ifo over the whole sample period (Jan. 1994 -Mar. 2002 and especially over horizons from six to twelve months, the ifo predicts better at shorter horizons (up to three months) and is superior to the ZEW indicator when a shorter sample (Jan. 1998 -Mar. 2002) is regarded. The PMI exhibits the worst forecasting performance and does not beat a naive reference model at all forecasting horizons.Taken both in-sample and out-of-sample results, we conclude that while the ifo, ZEW and PMI indicator lead the release of industrial production statistics in Germany by about six months, only ifo and ZEW perform well in out-of-sample forecasts (which might be due to the short sample period for the PMI). While the ifo indicator predicts better at short horizons of up to three months, the ZEW ...

On the Forecasting Properties of the Alternative Leading Indicators for the German GDP: Recent Evidence

“…In doing so, our paper is related to rather extensive literature that assesses the forecasting properties of various leading indicators for Germany represented by Benner and Meier (2003), Breitung and Jagodzinski (2001), Dreger and Schumacher (2005), Fritsche and Stephan (2000), Hüfner and Schröder (2002a), and Hüfner and Schröder (2002b) for growth rates of German industrial production, and Dreger and Schumacher (2004), Hinze (2003), Langmantel (1999), Mittnik and Zadrozny (2004), and Schumacher (2005) for growth rates of German real GDP.…”

The early news catches the attention: On the relative price impact of similar economic indicators

“…Both indicators were designed as leading indicators for changes in real economic activity. In line with the indicators' self-declared benchmark of measuring future economic activity, Huefner and Schroeder (2002) find that both indicators are useful to predict the year-by-year growth rate of German industrial production. Therefore, we follow Huefner and Schroeder (2002) and benchmark both indicators against the growth rate of industrial production.…”

Do Ifo Indicators Help Explain Revisions in German Industrial Production?

“…Rather than focusing on the forecasting ability of Ifo Business Survey indicators, as is done for instance by Langmantel (1999), Fritsche and Stephan (2002) and Hüfner and Schröder (2002), our paper deals with the strength of some of these indicators in explaining revisions of growth rates of German industrial production. We carry out a real-time analysis and examine vintages of data series on industrial production.…”