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,
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.
Artikel Lesen
Gebietsstands-Transformation Deutschland
Schlüsselbrücken zur Gebietsstands-Transformation in Deutschland Der Staat besitzt die Möglichkeit, innerhalb seiner Staatsgrenzen die ursprüngliche räumliche Struktur seiner…
Zur Seite
Daten
Schlüsselbrücken zur Gebietsstands-Transformation in Deutschland – Daten Zur Demonstration, in welcher Form die Daten aufbereitet und angeboten werden, stellen wir aus den…
Zur Seite
PhD Graduates Financial Markets
PhD Graduates of the Department of Financial Markets Eleonora Sfrappini: "Four Essays on Banking, Climate Risks and Financial Regulation" (2024) Willam McShane: "The Competitive…
Zur Seite
Macroeconomic Effects from Sovereign Risk vs. Knightian Uncertainty
Ruben Staffa
IWH Discussion Papers,
Nr. 27,
2023
Abstract
This paper compares macroeconomic effects of Knightian uncertainty and risk using policy shocks for the case of Italy. Drawing on the ambiguity literature, I use changes in the bid-ask spread and mid-price of government bonds as distinct measures for uncertainty and risk. The identification exploits the quasi-pessimistic behavior under ambiguity-aversion and the dealer market structure of government bond markets, where dealers must quote both sides of the market. If uncertainty increases, ambiguity-averse dealers will quasi-pessimistically quote higher ask and lower bid prices – increasing the bid-ask spread. In contrast, a pure change in risk shifts the risk-compensating discount factor which is well approximated by the change in bond mid-prices. I evaluate economic effects of the two measures within an instrumental variable local projection framework. The main findings are threefold. First, the resulting shock time series for uncertainty and risk are uncorrelated with each other at the intraday level, however, upon aggregation to monthly level the measures become correlated. Second, uncertainty is an important driver of economic aggregates. Third, macroeconomic effects of risk and uncertainty are similar, except for the response of prices. While sovereign risk raises inflation, uncertainty suppresses price growth – a result which is in line with increased price rigidity under ambiguity.
Artikel Lesen
Das IWH auf der Jahrestagung des Vereins für Socialpolitik 2019 "30 Jahre Mauerfall" - Demokratie und Marktwirtschaft
IWH-BROWN-BAG-PANEL "Ost-West-Produktivitätslücke: Ursachen und Folgen" Ostdeutschlands Wirtschaft konnte anfänglich ihre Produktivität gegenüber den westdeutschen Verhältnissen…
Zur Seite
Department Profiles
Research Profiles of the IWH Departments All doctoral students are allocated to one of the four research departments (Financial Markets – Laws, Regulations and Factor Markets –…
Zur Seite
Projekte
Unsere Projekte 07.2022 ‐ 12.2026 Evaluierung des InvKG und des Bundesprogrammes STARK Bundesministerium für Wirtschaft und Klimaschutz (BMWK) Im Auftrag des Bundesministeriums…
Zur Seite
The Promise and Peril of Entrepreneurship
Robert W. Fairlie, Zachary Kroff, Javier Miranda, Nikolas Zolas
MIT Press,
2023
Abstract
Startups create jobs and power economic growth. That's an article of faith in the United States—but, as The Promise and Peril of Entrepreneurship reveals, our faith may be built on shaky ground. Economists Robert Fairlie, Zachary Kroff, Javier Miranda, and Nikolas Zolas—working with Census Bureau microdata—have developed a new data set, the Comprehensive Startup Panel, that tracks job creation and the survival of every startup in the country. In doing so, they recalibrate our understanding of how startups behave in the US economy. Specifically, their work seeks to answer three critical questions: How many jobs does each entrepreneur create? Do those jobs disappear quickly? And how long do entrepreneurial enterprises survive?
Artikel Lesen
Why Is the Roy-Borjas Model Unable to Predict International Migrant Selection on Education? Evidence from Urban and Rural Mexico
Stefan Leopold, Jens Ruhose, Simon Wiederhold
Abstract
The Roy-Borjas model predicts that international migrants are less educated than nonmigrants because the returns to education are generally higher in developing (migrant-sending) than in developed (migrant-receiving) countries. However, empirical evidence often shows the opposite. Using the case of Mexico-U.S. migration, we show that this inconsistency between predictions and empirical evidence can be resolved when the human capital of migrants is assessed using a two-dimensional measure of occupational skills rather than by educational attainment. Thus, focusing on a single skill dimension when investigating migrant selection can lead to misleading conclusions about the underlying economic incentives and behavioral models of migration.
Artikel Lesen
