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
Is Risk the Fuel of the Business Cycle? Financial Frictions and Oil Market Disturbances
Christoph Schult
IWH Discussion Papers,
No. 4,
2024
Abstract
I estimate a dynamic stochastic general equilibrium (DSGE) model for the United States that incorporates oil market shocks and risk shocks working through credit market frictions. The findings of this analysis indicate that risk shocks play a crucial role during the Great Recession and the Dot-Com bubble but not during other economic downturns. Credit market frictions do not amplify persistent oil market shocks. This result holds as long as entry and exit rates of entrepreneurs are independent of the business cycle.
Read article
A Congestion Theory of Unemployment Fluctuations
Yusuf Mercan, Benjamin Schoefer, Petr Sedláček
American Economic Journal: Macroeconomics,
No. 1,
2024
Abstract
We propose a theory of unemployment fluctuations in which newhires and incumbentworkers are imperfect substitutes. Hence, attempts to hire away the unemployed during recessions diminish the marginal product of new hires, discouraging job creation. This single feature achieves a ten-fold increase in the volatility of hiring in an otherwise standard search model, produces a realistic Beveridge curve despite countercyclical separations, and explains 30–40% of U.S. unemployment fluctuations. Additionally, it explains the excess procyclicality of new hires’ wages, the cyclical labor wedge, countercyclical earnings losses from job displacement, and the limited steady-state effects of unemployment insurance.
Read article
IWH-Flash-Indikator IV. Quartal 2023 und I. Quartal 2024
Katja Heinisch, Oliver Holtemöller, Axel Lindner, Birgit Schultz
IWH-Flash-Indikator,
No. 4,
2023
Abstract
Im dritten Quartal 2023 sank die Wirtschaftsleistung in Deutschland leicht um 0,1%, das schwache Plus aus dem Vorquartal wurde damit wieder abgeschmolzen. Insbesondere nahmen die Konsumausgaben der privaten Haushalte weiter ab. Das dürfte nicht zuletzt der immer noch recht kräftigen Inflation bei nur moderat steigenden Haushaltsbudgets geschuldet sein. Auch ist die Verunsicherung der privaten Haushalte nach wie vor groß, etwa bezüglich der Finanzierbarkeit der künftig notwendigen Klimaschutzmaßnahmen oder bezüglich der mittelfristigen Wirtschaftsaussichten in Deutschland.
Zudem haben sich die geopolitischen Risiken mit dem Ausbruch kriegerischer Handlungen im Nahen Osten noch einmal erhöht. Auch wenn für das vierte Quartal 2023 aufgrund wieder etwas steigender Realeinkommen ein kleiner Zuwachs der Produktion in Deutschland zu erwarten ist, lässt der Aufschwung auf sich warten. Das Bruttoinlandsprodukt (BIP) dürfte laut IWH-Flash-Indikator im vierten Quartal 2023 sowie im ersten Quartal 2024 jeweils um 0,2% steigen (vgl. Abbildung 1).
Read article
Understanding Post-Covid Inflation Dynamics
Martín Harding, Jesper Lindé, Mathias Trabandt
Journal of Monetary Economics,
November
2023
Abstract
We propose a macroeconomic model with a nonlinear Phillips curve that has a flat slope when inflationary pressures are subdued and steepens when inflationary pressures are elevated. The nonlinear Phillips curve in our model arises due to a quasi-kinked demand schedule for goods produced by firms. Our model can jointly account for the modest decline in inflation during the Great Recession and the surge in inflation during the post-COVID period. Because our model implies a stronger transmission of shocks when inflation is high, it generates conditional heteroskedasticity in inflation and inflation risk. Hence, our model can generate more sizeable inflation surges due to cost-push and demand shocks than a standard linearized model. Finally, our model implies that the central bank faces a more severe trade-off between inflation and output stabilization when inflation is elevated.
Read article
Media Response
Media Response October 2024 Oliver Holtemöller: Weshalb junge Menschen Ostdeutschland verlassen in: ARD tagesschau, 03.10.2024 Oliver Holtemöller: Mecklenburg-Vorpommerns Stärke…
See page
People
People Doctoral Students PhD Representatives Alumni Supervisors Lecturers Coordinators Doctoral Students Afroza Alam (Supervisor: Reint Gropp ) Julian Andres Diaz Acosta…
See page
East Germany
The Nasty Gap 30 years after unification: Why East Germany is still 20% poorer than the West Dossier In a nutshell The East German economic convergence process is hardly…
See page
Archive
Media Response Archive 2021 2020 2019 2018 2017 2016 December 2021 IWH: Ausblick auf Wirtschaftsjahr 2022 in Sachsen mit Bezug auf IWH-Prognose zu Ostdeutschland: "Warum Sachsens…
See page
Joint Economic Forecast
Joint Economic Forecast The joint economic forecast is an instrument for evaluating the overall economic situation and development in Germany, the euro area and the rest of the…
See page