Nowcasting and forecasting economic growth in the euro area using principal components

Many empirical studies show that factor models have a relatively high forecast compared to alternative short-term forecasting models. These empirical findings have been established for different data sets and for different forecast horizons. However, choosing the appropriate factor model specification is still a topic of ongoing debate. Moreover, the forecast performance during the recent financial crisis is not well documented. In this study we investigate these two issues in depth. We empirically test the forecast performance of three factor model approaches and report our findings in an extended empirical out-of-sample forecasting competition for the euro area and its five largest countries over the period 1992-2012. Besides, we introduce two extensions to the existing factor models to make them more suited for real-time forecasting. We show that the factor models were able to systematically beat the benchmark autoregressive model, both before as well as during the financial crisis. The recently proposed collapsed dynamic factor model shows the highest forecast accuracy for the euro area and the majority of countries we analyzed. The improvement against the benchmark model can range up to 77%, depending on the country and forecast horizon.
 
Keywords: Factor models, Principal component analysis, Forecasting, Kalman filter, State space method, Publication lag, Mixed frequency.

Working paper no. 415