 Are GDP Revisions Predictable? Evidence for SwitzerlandNo 4219, EcoMod2012 from EcoMod Abstract: This study presents a model that delivers more accurate forecasts of the revised rather initial estimates of the quarterly GDP growth rate in Switzerland during the period of the recent financial crisis. The key explanation to our findings is that our model, capitalizing on the information contained in the Business Tendency Surveys, is able to predict future revisions of the initial estimates. Our findings imply that there seems to be a scope for improvement of how preliminary estimates of the quarterly GDP growth rate are produced in Switzerland.A mixed-frequency small-scale dynamic factor model is used for forecasting. The model parameters are estimated by means of the Kalman filter. This model combines GDP growth available at a quarterly frequency and monthly indicators based on firms' surveys. The different publication lags are also taken into account.In this paper we constructed a small-scale mixed-frequency dynamic factor model using data for Switzerland. The factor model combines the quarterly GDP growth rate and the monthly survey indicators. We evaluate the forecasting performance of the model during the period of the recent financial crisis when accurate information on the current stance of the economy is especially in high demand. We demonstrate that this factor model produces more accurate forecasts than the alternative benchmark models such as a random-walk model and a first-order autoregressive model. More importantly, the factor model produces more accurate forecasts of the revised rather than first-published estimates of the GDP growth rate. We demonstrate that this remarkable finding could be explained by the fact that the factor model is useful for predicting not only directions of future GDP revisions but also their magnitude, at least during the period under scrutiny. We conclude that there seems to be a scope for improvement of how estimates of the GDP growth rate are produced in Switzerland: in particularly, in the direction of reducing volatility of subsequent revisions. Keywords: Switzerland; Business cycles; Forecasting; nowcasting (search for similar items in EconPapers) Downloads: (external link) Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:ekd:002672:4219 Access Statistics for this paper More papers in EcoMod2012 from EcoMod Contact information at EDIRC. |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to econpapers@oru.se.
EconPapers is hosted by the
School of Business at Örebro University.