CompNet
Welcome to CompNet CompNet Data Provider Forum & TSI Workshop The upcoming CompNet Data Providers Forum and TSI Workshop, taking place in Amsterdam on November 18-19, 2024,…
See page
Alumni
IWH Alumni The IWH maintains contact with its former employees worldwide. We involve our alumni in our work and keep them informed, for example, with a newsletter. We also plan…
See page
Centre for Business and Productivity Dynamics
Centre for Business and Productivity Dynamics (IWH-CBPD) The Centre for Business and Productivity Dynamics (CBPD) was founded in January 2025 and works with policy and research…
See page
CompNet Database
The CompNet Competitiveness Database The Competitiveness Research Network (CompNet) is a forum for high level research and policy analysis in the areas of competitiveness and…
See page
Internationalisation
Internationalisation The Leibniz Institute for Economic Research Halle (IWH) is responsible for economic research and economic policy advice on a scientific basis. The institute…
See page
International Banking Library
International Banking Library The International Banking Library (IBL) is a web-based platform for the exchange of research on cross-border banking. It provides access to data…
See page
IWH FDI Micro Database
IWH FDI Micro Database The IWH FDI Micro Database (FDI = Foreign Direct Investment) comprises a total population of affiliates of multinational enterprises (MNEs) in selected…
See page
Reports of the European Forecasting Network (EFN)
Reports of the European Forecasting Network (EFN) The European Forecasting Network (EFN) was a group of macroeconomic experts from different European research institutions (such…
See page
Forecasting Economic Activity Using a Neural Network in Uncertain Times: Monte Carlo Evidence and Application to the
German GDP
Oliver Holtemöller, Boris Kozyrev
IWH Discussion Papers,
No. 6,
2024
Abstract
In this study, we analyzed the forecasting and nowcasting performance of a generalized regression neural network (GRNN). We provide evidence from Monte Carlo simulations for the relative forecast performance of GRNN depending on the data-generating process. We show that GRNN outperforms an autoregressive benchmark model in many practically relevant cases. Then, we applied GRNN to forecast quarterly German GDP growth by extending univariate GRNN to multivariate and mixed-frequency settings. We could distinguish between “normal” times and situations where the time-series behavior is very different from “normal” times such as during the COVID-19 recession and recovery. GRNN was superior in terms of root mean forecast errors compared to an autoregressive model and to more sophisticated approaches such as dynamic factor models if applied appropriately.
Read article