Neural networks in developing and validating Expected Shortfall estimation models

Szerzők

  • Máté Uzsoki University of Pécs - Faculty Of Business And Economics
  • Gyöngyi Bugár University of Pécs - Faculty of Business and Economics
  • Balázs Kovács University of Pécs - Faculty of Business and Economics

Kulcsszavak:

Risk management, Expected shortfall, Modelling market risk, Artificial intelligence

Absztrakt

The new Basel III standards establish a revised framework for determining the capital charge for market
risk in internal models, transitioning from Value-at-Risk (VaR) to Expected Shortfall (ES). The latter is
a risk measure that more effectively captures tail risk and exhibits several advantageous properties, such
as coherence. Numerous models have been proposed in the literature for estimating ES, ranging from
historical simulation and parametric statistical models to quantile regression and Generalized
Autoregressive Conditional Heteroscedasticity (GARCH) methodology. Our goal is to develop a multilevel
adaptive neural network which is capable to provide significantly improved ES estimates compared
to a benchmark model, selected for its previous superior prediction performance compared to other
models.

Szerző életrajzok

Máté Uzsoki, University of Pécs - Faculty Of Business And Economics

Maté Uzsoki completed his studies at the Faculty of Business and Economics of the University of Pécs, the Budapest University of Technology and Economics, and the Oxford School between 2001 and 2013. After obtaining two BSc degrees and an MBA, he earned his PhD from the Doctoral School of Business Administration at the University of Pécs in 2022. His research focuses on the measurement of market risk in the field of quantitative finance and economic mathematics using neural networks and statistical models (MTMT identification number: 10051008). With over 10 years of professional experience gained at Gémklub, Aion Next Kft., Microsoft, and Licquid, he founded Serpentarius Software Ltd. in 2013, where he currently works as CEO. Since February 2024, he has been the Chief Operating Officer of SunDell NyRt., a real estate transaction-focused company with a consolidated balance sheet total of 60 billion forints. Since 2022, he has been the chair of the Advisory Board of the Department of Management Science at the Faculty of Business and Economics of the University of Pécs and a member of the Faculty Advisory Board

Gyöngyi Bugár, University of Pécs - Faculty of Business and Economics

Gyöngyi Bugár graduated from the Eötvös Loránd University in 1988 with an MSc in Mathematics and Physics, and in 1993 from the Faculty of Economics of the University of Pécs with an MSc in Economics and Business. In 1996, she received her Dr Univ title, followed by a PhD at the Faculty of Economics of the University of Pécs in 2001, and a habilitation in 2011.  In 2002, she has been appointed as an associate professor at the Faculty, and as of 1 July 2023, she is the head of the Department of Management Science. Her teaching activities include undergraduate, master's and doctoral programmes both in Hungarian and English, where she teaches courses in business mathematics, quantitative methods, optimum calculus and research methodology. Her main research interests cover quantitative risk management and international portfolio diversification. She has received several awards for her research, among them she has won the prestigious Bolyai Fellowship from the Hungarian Academy of Sciences twice and the Best Paper Award four times at various international conferences.

Balázs Kovács, University of Pécs - Faculty of Business and Economics

Balázs Kovács graduated in 2010 from the Faculty of Economics of the University of Pécs, majoring in Financial Instruments and Institutions, as well as a second major in German-Hungarian translation. He has been a student at the Doctoral School of Economics since the fall of 2010. His research area is the data and text mining modeling possibilities of foreign exchange rates. In 2011, he won the Andras Felkai Academic Award, which is awarded by Citibank Hungary and the Foundation for Developing Financial Culture every year to support the training of domestic financial professionals. From September 2012 to the end of 2017, he worked as an assistant lecturer at the Department of Economic Methodology, then at the Department of Management Science, and then as an assistant professor from January 2018. He teaches several methodological courses at our faculty within the fields of mathematics and IT. In addition to teaching, he was a professor member of the Student Benefit and Dormitory Committee of the Faculty of Economics until its termination, and participated as a predoctoral researcher in the project SROP-4.2.2.C-11/1/KONV-2012-0005, Well-being in the Information Society. He admires free and open source solutions and encourages everyone to prefer them. 

UPFBE WP Series 202606

Downloads

Megjelent

2026-11-02

Hogyan kell idézni

Uzsoki, M., Bugár, G., & Kovács, B. (2026). Neural networks in developing and validating Expected Shortfall estimation models. PTE KTK Műhelytanulmányok, 2026(6). Elérés forrás https://journals.lib.pte.hu/index.php/workingpapers/article/view/9785

Folyóirat szám

Évf. 2026 szám 6: Neural networks in developing and validating Expected Shortfall estimation models

Rovat

Műhelytanulmányok

Categories