Forecast Combination and Interpretability Using Random Subspace
Boris Kozyrev
IWH Discussion Papers,
No. 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.
Read article
Early Child Care, Maternal Labor Supply, and Gender Equality: A Randomized Controlled Trial?
Henning Hermes, Marina Krauß, Philipp Lergetporer, Frauke Peter, Simon Wiederhold
IWH Discussion Papers,
No. 14,
2024
Abstract
We provide experimental evidence that enabling access to universal early child care increases maternal labor supply and promotes gender equality among families with lower socioeconomic status (SES). Our intervention offers information and customized help with child care applications, leading to a boost in child care enrollment among lower-SES families. 18 months after the intervention, we find substantial increases in maternal full-time employment (+160%), maternal earnings (+22%), and household income (+10%). Intriguingly, the positive employment effects are not only driven by extended hours at child care centers, but also by an increase in care hours by fathers. Gender equality also benefits more broadly from better access to child care: The treatment improves a gender equality index that combines information on intra-household division of working hours, care hours, and earnings by 40% of a standard deviation, with significant increases in each dimension. For higher-SES families, we consistently observe negligible, insignificant treatment effects.
Read article
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
Microdata for Economic Research in Europe: Challenges and Proposals
Eric Bartelsman, Marco Matani, Filippo di Mauro, Sergio Inferrera, Ugo Panizza, Michael Polder
CEPR,
No. 18640,
2023
Abstract
While access to high-quality microdata is essential for economic research and policy evaluation, effective access to such data remains limited in Europe. It varies from country to country, with uneven information on access procedures. This is a major obstacle to social science research, including research on European competitiveness and the effects of climate change, inequality, globalization, and digitalization. The objective of this paper, which is based on a brainstorming exercise coordinated by CEPR and CompNet, is to assess the status quo and discuss a series of proposals for improving access to Microdata for economic research. We underline the need for developing the relevant tools for extended access to and use of European business statistics microdata. Building such tools entails both establishing the requested microdata and creating a body facilitating cross-country access to the established databases with harmonized content.
Read article
Media Response
Media Response November 2024 Steffen Müller: Firmaer er oftere i problemer in: Flensborg Avis, 22.11.2024 Steffen Müller: Zahl der Insolvenzen steigt in: Süddeutsche.de,…
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
Frühkindliche Betreuung erhöht den Arbeitsmarkterfolg von Müttern ohne Abitur
Henning Hermes, Marina Krauß, Philipp Lergetporer, Frauke Peter, Simon Wiederhold
Wirtschaft im Wandel,
No. 3,
2023
Abstract
In den meisten Ländern wirkt sich die Geburt eines Kindes negativ auf den Arbeitsmarkterfolg von Müttern aus, insbesondere bei Müttern mit niedrigerem Schulabschluss. In diesem Beitrag werden die Ergebnisse eines Feldexperiments in Deutschland vorgestellt, in dem Familien bei der Bewerbung für einen Platz in einer Kindertagesstätte (Kita) unterstützt wurden. Die Ergebnisse zeigen, dass der verbesserte Zugang zu frühkindlicher Betreuung die Wahrscheinlichkeit erhöht, dass Mütter ohne Abitur in Vollzeit arbeiten, und deren Haushaltseinkommen steigert. Um den Arbeitsmarkterfolg von Müttern zu verbessern, sollte die Politik den Zugang zu frühkindlicher Betreuung erleichtern und die Zahl der Kita-Plätze noch weiter erhöhen.
Read article
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.
Read article
Males Should Mail? Gender Discrimination in Access to Childcare
Henning Hermes, Philipp Lergetporer, Fabian Mierisch, Frauke Peter, Simon Wiederhold
American Economic Association Papers and Proceedings,
May
2023
Abstract
This study investigates discrimination against women when searching and applying for childcare in a nationwide field experiment. We send emails from fictitious parents to 9,313 childcare centers in Germany inquiring about access to childcare. We randomize whether the email is sent by the child's mother or father. Our results show that women receive shorter and less positive responses than men. The probability of receiving a response does not differ by gender, highlighting the importance of going beyond response rates to detect discrimination. We provide suggestive evidence that regional differences in gender discrimination are related to gender norms.
Read article
Early Child Care and Labor Supply of Lower-SES Mothers: A Randomized Controlled Trial
Henning Hermes, Marina Krauss, Philipp Lergetporer, Frauke Peter, Simon Wiederhold
CESifo Working Paper,
No. 10178,
2022
Abstract
We present experimental evidence that enabling access to universal early child care for families with lower socioeconomic status (SES) increases maternal labor supply. Our intervention provides families with customized help for child care applications, resulting in a large increase in enrollment among lower-SES families. The treatment increases lower-SES mothers' full-time employment rates by 9 percentage points (+160%), household income by 10%, and mothers' earnings by 22%. The effect on full-time employment is largely driven by increased care hours provided by child care centers and fathers. Overall, the treatment substantially improves intra-household gender equality in terms of child care duties and earnings.
Read article
