Forecasting Czech GDP Using Mixed-Frequency Data Models

In this paper we use a battery of various mixed-frequency data models to forecast Czech GDP. The models employed are mixed-frequency vector autoregressions, mixed-data sampling models, and the dynamic factor model. Using a dataset of historical vintages of unrevised macroeconomic and financial data, we evaluate the performance of these models over the 2005–2012 period and compare them with the Czech National Bank’s macroeconomic forecasts. The results suggest that for shorter forecasting horizons the accuracy of the dynamic factor model is comparable to the CNB forecasts. At longer horizons, mixed-frequency vector autoregressions are able to perform similarly or slightly better than the CNB forecasts. Furthermore, moving away from point forecasts, we also explore the potential of density forecasts from Bayesian mixed-frequency vector autoregressions.