Assessing Predictive Content of the KOF Barometer in Real Time

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may AbstractWe investigate whether the KOF Barometer-a leading indicator regularly released by the KOF Swiss Economic Institute-can be useful for short-term out-of-sample prediction of year-on-year quarterly real GDP growth rates in Switzerland. We find that the KOF Barometer appears to be useful for prediction of GDP growth rates. Even the earliest forecasts, made seven months ahead of the first official GDP estimate, allow us to predict GDP growth rates more accurately than forecasts based on an univariate autoregressive model. At every subsequent forecast round as new monthly releases of the KOF Barometer become available we observe a steady increase in forecast accuracy.

Assessing the Real-Time Informational Content of Macroeconomic Data Releases for Now-/Forecasting GDP: Evidence for Switzerland

“…First, to the best of our knowledge, it represents the first attempt to predict the growth rate of the Swiss GDP for the current quarter (nowcast) as well as for the next quarter (forecast) using a large-scale factor model. The previous research either used a single leading indicator model (Müller and Köberl, 2008) or it used a leading indicator (KOF-Barometer) extracted from a small-scale static factor model (Graff, 2009;Siliverstovs, 2010) for predicting GDP growth rates in Switzerland. Our study also distinguishes itself from these studies in that we choose the quarter-onquarter seasonally adjusted GDP growth rates as a target prediction variable rather than the year-on-year quarterly GDP growth rate that these earlier studies aimed to predict.…”

Assessing the Real-Time Informational Content of Macroeconomic Data Releases for Now-/Forecasting GDP: Evidence for Switzerland

“…First, to the best of our knowledge, it represents the first attempt to predict the growth rate of the Swiss GDP for the current quarter (nowcast) as well as for the next quarter (forecast) using a large-scale factor model. The previous research either used a single leading indicator model (Müller and Köberl, 2008) or it used a leading indicator (KOF-Barometer) extracted from a small-scale static factor model (Graff, 2009;Siliverstovs, 2010) for predicting GDP growth rates in Switzerland. Our study also distinguishes itself from these studies in that we choose the quarter-onquarter seasonally adjusted GDP growth rates as a target prediction variable rather than the year-on-year quarterly GDP growth rate that these earlier studies aimed to predict.…”