Dynamic forecast averaging of macroeconomic models

Investor logo
Project Identification
Project Period
1/2017 - 12/2019
Investor / Pogramme / Project type
Czech Science Foundation
MU Faculty or unit
Faculty of Economics and Administration

The forecasting literature has gained new momentum in the last decade, when newly developed DSGE (Dynamic Stochastic General Equilibrium) models started to forecast so well that their forecast performance began to compete with purely empirical models. It has been shown that averaging across DSGE and time series models may achieve better forecasting properties than the single best specification across all model classes. The forecasting performance of DSGE models varies over time. Still, we can make use of these estimates to improve the forecasting accuracy. We aim to address the shortcomings of the prevailing methodology by combining a large set of empirical and theoretical models to forecast macroeconomic quantities in all major economies. As compared to the literature, the comparatively larger model space calls for more efficient algorithms that efficiently combine the predictions obtained from individual models. We plan to introduce new combination schemes that incorporate several stylized facts commonly observed in the macroeconomic forecasting literature.


Total number of publications: 9

You are running an old browser version. We recommend updating your browser to its latest version.