Interview with Prof. Dr. Daniel Rösch

Professor Dr. Daniel Rösch holds the chair for statistics and risk management at the University of Regensburg. Before moving to the University of Regensburg in 2013, he was professor of finance from 2007 to 2013 and director of the institute for Banking in Finance at Leibniz Universität Hannover. From 2006 to 2011 he was a visiting researcher at the University of Melbourne.

He has been a visiting professor at the University of Technology in Sydney since 2011. His research interests include banking, risk management, credit risk analysis, financial regulation and supervision, data science, machine learning and real estate finance. He has published numerous articles in leading international trade journals, received several awards and honors and regularly gives lectures at important international conferences. As a service provider, he was former President of the German Finance Association, co-founder and member of the Board of Directors of the Hanover Financial Center and deputy managing director of the Finance and Financial Institutions working group of Operations Research Society. He is also an editorial member of the Journal of Risk Model Validation. Professor Rösch has worked on joint research projects with financial institutions and regulators such as the Deutsche Bundesbank.

He is currently participating in a long term research grant with scientists from the University of Technology in Sydney, which is supported by the Australian Center for Internatinal finance and Regulation.

Prof. Dr. Daniel Rösch

Chair of Statistics and Risk Management

I hold the Chair for Statistics and Risk Management at the Faculty of Economics.

We teach on the one hand in the field of statistics, data science and machine learning with a view to economic issues, on the other hand in the area of financial risk management, i.e. the measurement, analysis and forecasting of risks e.g. of banks and insurance companies.

We show students how they can solve problems and questions in the field of economics using methods from statistics, data science and machine learning. E.g. banks and insurance companies have to forecast future financial risks. To do this, they use statistical methods and machine learning and we equip the students with the skills to use these methods correctly and sensibly in practice.

We use a mixture of theoretical basics and current practical applications with real data and computer programs in Python, R and SAS. Since our chair is actively researching, we can always incorporate current and relevant research results. Students acquire the necessary prior knowledge in our basic courses Statistics 1 and 2 and in various courses on finance.

Methodological competence, ability to abstract, assess and communicate.

Because the amount of data is getting bigger and the computing power increases enormously, data science and automated processes have become indispensable as decision support systems in practice.

Analogous to teaching, our research focuses on methods for the empirical analysis of data for the purpose of forecasting, especially the forecasting of financial risks, in order to better understand reality and to be able to act and control it better.

We are currently using machine learning methods (especially neural networks) in various projects in order to be able to better predict risks that banks suffer from loan defaults. This is extremely important and relevant in practice, especially in times of the corona crisis.

Studies, doctorate, habilitation, professorship… My entire career has been and is determined by statistical and machine learning methods and their use in research and practice.

Basically, of course, easy. But it would be boring if there were not always new challenges…

That these methods have great potential for decision support. I don’t think you can replace people as decision-makers, but you can provide them with powerful tools for making decisions about data science. A current example: In the “Corona Crisis” decisions were / are made by politicians, the future scope of which is still completely unclear to all of us. In addition, the decisions are based on thin data. It would have been extremely helpful (but unfortunately not possible) to have better information and data as a basis.

I am happy about all initiatives that not only exist “on paper” but also generate real added value. I see this here too.


Mr. Rösch, thank you for taking the time to answer these questions. We wish you a nice day!

Last edited on