Third party funded

AI*Teaching 2026

The project ‘Use of Artificial Intelligence in Python Programming’ by Prof. Dr Eva Lütkebohmert-Holtz and Marcus Rockel is being funded as part of the ‘AI*teaching 2026’ ideas competition.

Since 2023, the “Python Course” lecture has been equipping students with practical skills in software development and data engineering. In light of the profound transformation brought about by AI, the course will expand in the winter semester of 2026/27 to encourage the reflective use of AI-supported tools in programming. Through practice-oriented scenarios, traditional programming tasks will be integrated with contemporary automation and assistance systems, facilitating the development of more efficient and higher-quality solutions.

https://uni-freiburg.de/ideenwettbewerb-kilehre-2026-ermoeglicht-foerderung-von-15-projekten/

DFG-Project

Utilization of market information for option valuation, calibration, and portfolio management

Since the seminal work of Black and Scholes (1973), financial modeling has become increasingly sophisticated in an effort to capture complex market dynamics more accurately. Nevertheless, exact price replication remains both impractical and undesirable, given the potential for model misspecification and the presence of inherent market uncertainties. Furthermore, advanced option pricing models often involve parameters that are difficult to estimate reliably, which can significantly impact derivative pricing, hedging strategies, and option portfolio management. These challenges give rise to fundamental questions about how to efficiently incorporate market information to quantify parameter uncertainty, achieve robust pricing and hedging of derivatives, and enable model-independent approaches to portfolio management.

This project tackles these issues from multiple perspectives, offering methodological innovations alongside practical insights for derivative valuation, model calibration, and risk and portfolio management.

MDB-Project

Measuring Concentration Risk in MDB Portfolios

Multilateral Development Banks (MDBs) play an important role in financing sustainable development, as they raise large volumes of funds on international capital markets to finance projects that mitigate climate change and promote social development and economic growth in developing countries. In this sense, MDB debt is increasingly relevant as a specialized sub-component of fixed income and can be considered as a sub-set of the broader social/green bond space. 

As non-profit supranational development institutions, MDBs are unregulated and therefore there are no generally agreed standards to orient stakeholders. Given their financial model, it is therefore essential for MDBs to maintain strong credit ratings, which are, however, highly dependent on how credit rating agencies (CRAs) assess MDBs’ capital adequacy.

An important risk component in MDBs’ loan portfolios is their exposure to single name concentration risk, which arises from the fact that MDBs’ development-related lending consists mainly of sovereign lending, so that their loan portfolios typically consist of a small number of borrowers. For instance, the World Bank’s International Bank for Reconstruction and Development (IBRD) had only 78 sovereign borrowers as of June 2022, while the regional MDBs have even fewer. This exposes them to a high degree to name concentration risk, i.e., to the undiversified idiosyncratic risk associated with the default of individual borrowers.

The leading methodology that is currently in use (e.g., in S&P (2018)) to account for name concentration risk in MDBs’ capital adequacy framework relies on the Granularity Adjustment (GA) developed in Gordy and Lütkebohmert (2013). However, the latter was originally designed for commercial banks, which typically hold much larger portfolios consisting of at least several hundred borrowers. 

The aim of the project was to evaluate the above concerns within a comprehensive quantitative study and to develop a new methodology for the quantification of undiversified idiosyncratic risk which is tailored for small and concentrated MDB portfolios. The results of the project are presented in the following two publications:

• Lütkebohmert, E. und Sester, J. (2025) Measuring Name Concentrations through Deep Learning, International Review of Financial Analysis 107: 104598.
• Lütkebohmert, E., Sester, J., Shen, H. (2025) Name Concentration Risk in Multilateral Development Banks’ Portfolios: Measurement and Capital Adequacy Implications, Global Finance Journal 67: 101154.

Eva Lütkebohmert presented the results of the project on various international conferences as well as at the 3rd MDBs & Credit Rating Agencies Roundtable hosted by the Inter-American Developement Bank in Washington D.C., on the side lines of the 2024 Spring Meetings of the International Monetary Fund and the World Bank Group, April 2024 (see also https://www.iadb.org/en/news/readout-third-mdbs-credit-rating-agencies-roundtable).

In October 2025, Standard & Poor’s revised its capital adequacy framework for Multilateral Lending Institutions, introducing significant enhancements. These include improved recognition of MDBs’ preferred creditor status, a revision of parameters within the single-name concentration risk adjustment, and updated criteria for hybrid capital. Some of these updates relate to the work carried out in this project. See also https://www.spglobal.com/ratings/en/regulatory/article/-/view/type/HTML/id/3457348

This project was funded by the MDB Challenge Fund. The MDB Challenge Fund is administered by New Venture Fund and supported by grants from the Bill & Melinda Gates Foundation, Open Society Foundations and the Rockefeller Foundation. For further information see also https://mdbchallenge.com/

DFG-Project

Interest‑rate markets after the 2007‑2008 financial crisis: analysis, modeling, and stress testing of multiple yield curves

The 2007‑2008 financial crisis triggered structural changes in interest‑rate markets. Today’s markets are characterized by tenor‑dependent yield curves that reflect different risk categories. This has important implications for the valuation of interest‑rate derivatives, portfolio allocation and risk management.

Within the DFG‑funded project, significant contributions have been made to the modelling of multiple yield curves and to the pricing of interest‑rate derivatives.

The following publications have resulted from the project:

• Brignone, R., Gonzato, L. und Lütkebohmert, E. (2023) Efficient quasi-Bayesian estimation of affine option pricing models using risk-neutral cumulants, Journal of Banking and Finance 148: 106745.
• Brignone, R., Gerhart, C. und Lütkebohmert, E. (2022) Arbitrage-free Nelson-Siegel model for multiple yield curves, Mathematics and Financial Economics 16:
  239-266.
• Eberlein, E., Gerhart, C. und Lütkebohmert, E. (2020) A multiple curve Lévy swap market model, Applied Mathematical Finance 27(5): 396-421.

FRIAS

The Freiburg Institute for Advanced Studies (FRIAS) is the university’s international research college. As an integral part of the University of Freiburg, the institute brings together the humanities and social sciences as well as medicine, natural, life and engineering sciences under one roof.

Fellowship 2018/2019

During the academic year 2018/19, Prof. Eva Lütkebohmert‑Holtz held the position of Internal Senior Fellow at the Freiburg Institute for Advanced Studies (FRIAS). During this time, she worked on a project on the topic of

“Instability in interbank markets.”

The collapse of the interbank market during the 2007‑2008 financial crisis had far‑reaching consequences for the entire financial system, and the subsequent recession imposed severe repercussions on the broader economy. Understanding the causes of the instabilities that were observed both during and after the crisis—and developing models that explicitly incorporate these risk factors—is therefore of high societal relevance and constitutes the primary objective of this research project.

In particular, the project investigates the economic determinants that drive the spreads on inter‑bank loans, which surged dramatically during the 2007‑2008 crisis and have remained at a significantly elevated level ever since.

In addition, this project aims to develop a new dynamic model approach to analyze the relationship between capital outflows from lenders and trading on the interbank market, as well as to identify systemic risks in financial networks.

Our findings have numerous practical implications, among others for the valuation and hedging of interest rate derivatives, the risk management of financial institutions, and the development of regulatory tools to safeguard financial market stability.

The following publications have resulted from this project:

• Feng, X., Lütkebohmert, E. und Xiao, Y. (2022) Wealth management products, banking competition, and stability: Evidence from China, Journal of Economic Dynamics and Control 137: 104346, 2022, https://doi.org/10.1016/j.jedc.2022.104346.
  
• Gerhart, C. und Lütkebohmert, E. (2020) Empirical analysis and forecasting of multiple yield curves, Insurance: Mathematics and Economics 95: 59-78.
  
• Gerhart, C., Lütkebohmert, E. und Weber, M. (2019) Robust forecasting of multiple yield curves. In Valenzuela, O., Rojas, F., Pomares, H., Rojas, I. (Eds.): Theory and Applications of Time Series Analysis, Springer Contributions to Statistics, pp. 187-202.
  
• Gerhart, C., Lütkebohmert, E. und Weber, M. (2018) Forecasting of multiple yield curves based on machine learning. Proceedings of the International Conference on Time Series and Forecasting 2018, Vol. 3, pp. 1483-1494.

Project group 2017/2018

During the academic year 2017/18, Prof. Eva Lütkebohmert‑Holtz participated in a project group on “Model Risk” together with Prof. Patrick Dondl, JProf. Philipp Harms, and Prof. Thorsten Schmidt (Mathematical Institute) at the Freiburg Institute for Advanced Studies (FRIAS).

The aim of the project group was to develop new methods in financial market research and stochastic analysis that are both practically applicable and robust. In the natural sciences, interesting connections arise—for example when examining models in which a nonlinear dependence on random variables occurs.

The project group represents a first step in an agenda with high potential for future developments. Model risk plays a decisive role in all disciplines that employ mathematical models, e.g., medicine, physics, biology, computer science, etc. Its application to financial markets is of utmost societal relevance, underscoring the profound impact of recent financial crises. A key objective of this research team is to strengthen the linkage to real world applications.

The following activities were organized within this project group:

Workshop Robust Finance, FRIAS, May 16/17, 2018

Postbank Finance Award 2015

“The Postbank Finance Award was presented for the 12th time by the Postbank in Bonn. The award, endowed with €100,000, is Germany’s most prestigious university prize in finance. Since its inception in 2003, the Postbank Finance Award has been awarded annually with the aim of “understanding the future – shaping the future.” The competition seeks to promote innovative, academically sound answers to current finance‑related questions.

The prize aims to encourage students from all disciplines to engage with contemporary financial economic issues and to provide them with ideas and support for further study and career planning. Seventy percent of the prize money is allocated to the awarded universities for equipment and other resources.

The third place prize was won by Danjela Guxha, Mariia Markovych, Christiane Müller, and Daria Saulenko from the University of Freiburg. Together with Prof. Dr. Eva Lütkebohmert‑Holtz, the students developed a concept that shows how institutional investors can improve their returns by incorporating low‑liquidity securities into their portfolios.”

(https://www.finanzwelt.de/post/postbank-verleiht-den-12)

Instructional Development Award (IDA) 2016

The core of scientific education in finance and financial mathematics is the teaching of methods and the underlying theory at a high level and within a broad, general framework. In contrast, the problems actually tackled in companies are often more specialized because they have to incorporate many firm‑specific constraints. After graduating, students are confronted not only with these specialized issues but also with a multitude of new demands – working in interdisciplinary teams, operating under tight deadlines, coping with constantly changing tasks, and much more.

The aim of our project was to give students a glimpse of real‑world practice and to train them in dealing with these challenges. In addition, we sought to foster, as effectively as possible, the exchange and collaboration of students from economics and mathematics.

In cooperation with industry partners we selected a number of exciting, practice‑oriented problem statements and asked the students to solve them in interdisciplinary teams composed of economists and mathematicians, as independently as possible and within a prescribed time frame. A competitive atmosphere was created through regular presentations and peer‑evaluation sessions.

The implementation was realized through a new joint module for the M.Sc. programs in Economics, Business Administration and Mathematics, entitled “Finance from a Practice Perspective – Learning Motivated by Application.” The module includes a seminar with an intensive term‑paper component for which 6 ECTS are credited. Furthermore, a series of thematically relevant block courses were offered via external teaching assignments, rounding out the content and making it accessible to a wide audience.

DFG-Project

Modeling market, credit, and liquidity risks in fixed‑income markets

The assessment of financial risks has traditionally focused on market and credit risk, while liquidity risk has received far less attention. Yet, the scarcity of liquidity was one of the principal drivers of the 2007‑2009 financial crisis.

As part of the DFG-funded project, we have developed models that allow for the simultaneous assessment of multiple risk types. In particular, we have further developed the class of structural credit risk models so that, in addition to insolvency risks, liquidity risks can also be taken into account and potential feedback effects between the two risk classes can be mapped.

The following three publications resulted from this project:

• He, X.-Z., Lütkebohmert, E. und Xiao, Y. (2017) Rollover risk and credit risk under time-varying margin, Quantitative Finance. DOI: 10.1080/14697688.2016.1203071.
• Lütkebohmert, E., Oeltz, D. und Xiao, Y. (2017) Endogenous credit spreads and optimal debt financing structure in the presence of liquidity risk, European Financial Management. DOI: 10.1111/eufm.12089.
• Liang, G., Lütkebohmert, E. und Wei, W. (2015) Funding liquidity, debt tenor structure, and creditor’s belief: An exogenous dynamic debt run model, Mathematics and Financial Economics 9, pp. 271-302. DOI: 10.1007/s11579-015-0144-6.