Universität Regensburg

Daniel Rösch

Anschrift:

Leistungsprofil:

• Bank- und Versicherungswesen
• Risikomanagement, Kreditrisikomanagement
• Prudential Regulation
• Data Science, Data Analytics, Machine Learning, Statistics, Predictive Analytics

• Measurement and Validation of Bank Loan PD, LGD, EAD and Correlations
• Credit Risk Analytics
• Counterparty Credit Risk
• Deep Learning Credit Risk
• Deep Derivatives Calibration

Publikationen:

• Ralf Kellner, Maximilian Nagl, and Daniel Rösch. Opening the black box - quantile neural networks for loss given default prediction. Journal of Banking and Finance, 134:106334, 2022
• Daniel Rösch and Harald Scheule. Deep Credit Risk - Machine Learning in Python. Kindle Direct Publishing, 2020
• Jennifer Betz, Ralf Kellner, and Daniel Rösch. Time matters: How default resolution times impact final loss rates. Journal of the Royal Statistical Society, Series C, page 105996, 2021
• Yongwoong Lee, Harald Scheule, and Daniel Rösch. Systematic credit risk in securitised mortgage portfolios. Journal of Banking and Finance, 122:105996, 2021
• Jennifer Betz, Steffen Krüger, Ralf Kellner, and Daniel Rösch. Macroeconomic effects and frailties in the resolution of non-performing loans. Journal of Banking & Finance, 112:1–26, 2020
• Patrick Büchel, Michael Kratochwil, and Daniel Rösch. Computing valuation adjustments for counter- party credit risk using a modified supervisory approach. Review of Derivatives Research, 23(3):273–322, 2020
• Ralf Kellner and Daniel Rösch. A country specific point of view on international diversification. Journal of International Money and Finance, 98:102064, November 2019.
• Hung Do, Harald Scheule, and Daniel Rösch. Liquidity constraints, home equity and residential mortgage losses. Journal of Real Estate Finance and Economics, 2019.
• Rainer Jobst, Ralf Kellner, and Daniel Rösch. Bayesian loss given default estimation for european sovereign bonds. International Journal of Forecasting, 36:1073–1091, 2020
• Maximilian Nagl, Marius Pfeuffer, Matthias Fischer, and Daniel Rösch. Parameter estimation, bias correction and uncertainty quantification in the vasicek credit portfolio model. Journal of Risk, 2019

Kooperationsangebot für die Wirtschaft / Praxis:

• Datengetriebene Entscheidungsfindung (z. B. Business Analytics, Machine Learning, Artificial Intelligence): fachliche Expertise in der methodischen Aurichtung und der erfolgreichen Anwendung
• Angebot der Kooperation für Unternehmer*in/Unternehmen über Drittmittelprojekte oder Forschungsaufträge
• Wissenschaftliche Spitzenforschung zur Lösung praktischer Probleme

Bestehende Kooperationen: