Forecasting Economic Activity Using a Neural Network in Uncertain Times: Monte Carlo Evidence and Application to the German GDP
in: IWH Discussion Papers, Nr. 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.
Globalization, Productivity Growth, and Labor Compensation
in: IWH Discussion Papers, Nr. 7, 2022
Abstract
We analyze how changes in international trade integration affect productivity and the functional income distribution. To account for endogeneity, we construct a leaveout measure for international trade integration for country-industry pairs using international input-output tables. Our findings corroborate on the country-industry level that international trade integration increases productivity. Moreover, we show that both trade in intermediate inputs and trade in value added is associated with lower labor shares in emerging markets. For advanced countries, we document a positive effect of trade in value added on the labor share of income. Further, we show that the effects on productivity and labor share are heterogeneous across different sectors. Finally, we discuss the implications of our results for a possible throwback in international trade integration due to experiences from recent crises.
Exchange Rates and the Information Channel of Monetary Policy
in: IWH Discussion Papers, Nr. 17, 2020
Abstract
We disentangle the effects of monetary policy announcements on real economic variables into an interest rate shock component and a central bank information shock component. We identify both components using changes in interest rate futures and in exchange rates around monetary policy announcements. While the volatility of interest rate surprises declines around the Great Recession, the volatility of exchange rate changes increases. Making use of this heteroskedasticity, we estimate that a contractionary interest rate shock appreciates the dollar, increases the excess bond premium, and leads to a decline in prices and output, while a positive information shock appreciates the dollar, decreases prices and the excess bond premium, and increases output.