Forecast Combination and Interpretability Using Random Subspace
Boris Kozyrev
IWH Discussion Papers,
Nr. 21,
2024
Abstract
This paper investigates forecast aggregation via the random subspace regressions method (RSM) and explores the potential link between RSM and the Shapley value decomposition (SVD) using the US GDP growth rates. This technique combination enables handling high-dimensional data and reveals the relative importance of each individual forecast. First, it is possible to enhance forecasting performance in certain practical instances by randomly selecting smaller subsets of individual forecasts and obtaining a new set of predictions based on a regression-based weighting scheme. The optimal value of selected individual forecasts is also empirically studied. Then, a connection between RSM and SVD is proposed, enabling the examination of each individual forecast’s contribution to the final prediction, even when there is a large number of forecasts. This approach is model-agnostic (can be applied to any set of predictions) and facilitates understanding of how the aggregated prediction is obtained based on individual forecasts, which is crucial for decision-makers.
Artikel Lesen
Who Benefits from Place-based Policies? Evidence from Matched Employer-Employee Data
Philipp Grunau, Florian Hoffmann, Thomas Lemieux, Mirko Titze
IWH Discussion Papers,
Nr. 11,
2024
Abstract
We study the wage and employment effects of a German place-based policy using a research design that exploits conditionally exogenous EU-wide rules governing the program parameters at the regional level. The place-based program subsidizes investments to create jobs with a subsidy rate that varies across labor market regions. The analysis uses matched data on the universe of establishments and their employees, establishment-level panel data on program participation, and regional scores that generate spatial discontinuities in program eligibility and generosity. These rich data enable us to study the incidence of the place-based program on different groups of individuals. We find that the program helps establishments create jobs that disproportionately benefit younger and less-educated workers. Funded establishments increase their wages but, unlike employment, wage gains do not persist in the long run. Employment effects estimated at the local area level are slightly larger than establishment-level estimates, suggesting limited spillover effects. Using subsidy rates as an instrumental variable for actual subsidies indicates that it costs approximately EUR 25,000 to create a new job in the economically disadvantaged areas targeted by the program.
Artikel Lesen
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
Herding Behavior and Systemic Risk in Global Stock Markets
Iftekhar Hasan, Radu Tunaru, Davide Vioto
Journal of Empirical Finance,
September
2023
Abstract
This paper provides new evidence of herding due to non- and fundamental information in global equity markets. Using quantile regressions applied to daily data for 33 countries, we investigate herding during the Eurozone crisis, China’s market crash in 2015–2016, in the aftermath of the Brexit vote and during the Covid-19 Pandemic. We find significant evidence of herding driven by non-fundamental information in case of negative tail market conditions for most countries. This study also investigates the relationship between herding and systemic risk, suggesting that herding due to fundamentals increases when systemic risk increases more than when driven by non-fundamentals. Granger causality tests and Johansen’s vector error-correction model provide solid empirical evidence of a strong interrelationship between herding and systemic risk, entailing that herding behavior may be an ex-ante aspect of systemic risk, with a more relevant role played by herding based on fundamental information in increasing systemic risk.
Artikel Lesen
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
Micro Data on Robots from the IAB Establishment Panel
Verena Plümpe, Jens Stegmaier
Jahrbücher für Nationalökonomie und Statistik,
Nr. 3,
2023
Abstract
Micro-data on robots have been very sparse in Germany so far. Consequently, a dedicated section has been introduced in the IAB Establishment Panel 2019 that includes questions on the number and type of robots used. This article describes the background and development of the survey questions, provides information on the quality of the data, possible checks and steps of data preparation. The resulting data is aggregated on industry level and compared with the frequently used robot data by the International Federation of Robotics (IFR) which contains robot supplier information on aggregate robot stocks and deliveries.
Artikel Lesen
Household Indebtedness, Financial Frictions and the Transmission of Monetary Policy to Consumption: Evidence from China
Michael Funke, Xiang Li, Doudou Zhong
Emerging Markets Review,
June
2023
Abstract
This paper studies the impact of household indebtedness on the transmission of monetary policy to consumption using the Chinese household-level survey data. We employ a panel smooth transition regression model to investigate the non-linear role of indebtedness. We find that housing-related indebtedness weakens the monetary policy transmission, and this effect is non-linear as there is a much larger counteraction of consumption in response to monetary policy shocks when household indebtedness increases from a low level rather than from a high level. Moreover, the weakened monetary policy transmission from indebtedness is stronger in urban households than in rural households. This can be explained by the investment good characteristic of real estate in China.
Artikel Lesen
Uncovered Workers in Plants Covered by Collective Bargaining: Who Are They and How Do They Fare?
Boris Hirsch, Philipp Lentge, Claus Schnabel
British Journal of Industrial Relations,
Nr. 4,
2022
Abstract
Abstract In Germany, employers used to pay union members and non-members in a plant the same union wage in order to prevent workers from joining unions. Using recent administrative data, we investigate which workers in firms covered by collective bargaining agreements still individually benefit from these union agreements, which workers are not covered anymore and what this means for their wages. We show that about 9 per cent of workers in plants with collective agreements do not enjoy individual coverage (and thus the union wage) anymore. Econometric analyses with unconditional quantile regressions and firm-fixed-effects estimations demonstrate that not being individually covered by a collective agreement has serious wage implications for most workers. Low-wage non-union workers and those at low hierarchy levels particularly suffer since employers abstain from extending union wages to them in order to pay lower wages. This jeopardizes unions' goal of protecting all disadvantaged workers.
Artikel Lesen
Physical Climate Change and the Sovereign Risk of Emerging Economies
Hannes Böhm
Journal of Economic Structures,
2022
Abstract
I show that rising temperatures can detrimentally affect the sovereign creditworthiness of emerging economies. To this end, I collect long-term monthly temperature data of 54 emerging markets. I calculate a country’s temperature deviation from its historical average, which approximates present-day climate change trends. Running regressions from 1994m1 to 2018m12, I find that higher temperature anomalies lower sovereign bond performances (i.e., increase sovereign risk) significantly for countries that are warmer on average and have lower seasonality. The estimated magnitudes suggest that affected countries likely face significant increases in their sovereign borrowing costs if temperatures continue to rise due to climate change. However, results indicate that stronger institutions can make a country more resilient towards temperature shocks, which holds independent of a country’s climate.
Artikel Lesen
A Note on the Use of Syndicated Loan Data
Isabella Müller, Felix Noth, Lena Tonzer
IWH Discussion Papers,
Nr. 17,
2022
Abstract
Syndicated loan data provided by DealScan is an essential input in banking research. This data is rich enough to answer urging questions on bank lending, e.g., in the presence of financial shocks or climate change. However, many data options raise the question of how to choose the estimation sample. We employ a standard regression framework analyzing bank lending during the financial crisis of 2007/08 to study how conventional but varying usages of DealScan affect the estimates. The key finding is that the direction of coefficients remains relatively robust. However, statistical significance depends on the data and sampling choice and we provide guidelines for applied research.
Artikel Lesen
