Maximilian Nagl has been a research assistant at the Chair of Statistics and Risk Management at the University of Regensburg since 2018. Before that, he did his Bachelor’s and Master’s degree in Business Administration in Regensburg. His tasks as a lecturer include, on the one hand, holding exercises on the subjects of “Data Science & Machine Learning” and “Applied Data Science”, on the other hand, he is also responsible for supervising theses. His research focuses on the application of Bayesian statistics, machine learning in risk management and the assessment of credit risk.
Chair for Statistics and Risk Management
I have been working at Prof. Rösch’s Chair for Statistics and Risk Management since May 2018. My main focus is the quantification of credit risks.
My teaching activities include exercises for the courses “Applied Data Science” and “Data Science & Machine Learning”. I also supervise Bachelor theses, Master seminars and Master theses on the subject of machine learning and credit risk management.
In the exercises I supervise, we cover all common, but also advanced, methods of machine learning. In “Applied Data Science” we discuss both supervised and unsupervised methods. In “Data Science & Machine learning”, artificial neural networks take up a large part of the semester.
We discuss the methods both theoretically in order to provide the students with a sufficient basis, but also practically. Programming in Python is one of the central components in both courses and it is also deepened through case studies. Prior knowledge from our basic courses is always helpful here.
The students acquire a broad theoretical understanding of the methods and can apply this to real problems. Practical implementation in Python is also another important specialist skill.
I think that the topic of AI will accompany us in the next decades and students should deal with it as early as possible.
I am currently working on a combination of neural networks and classic advanced methods. The combination of both “worlds” could deliver promising results.
I am currently working with colleagues on a possibility to replace highly complex methods of valuing financial derivatives with artificial neural networks. On the one hand, this should improve the speed of the assessment, but it should also deliver more robust results.
My first contact came about as part of my research and was deepened primarily through exchanges with colleagues.
Question 10: Do you personally find it easy or difficult to deal with topics that are related to AI?
I find this area extremely exciting. Above all, one can adopt or solve problems / solutions from classic statistics.
The most fascinating thing are the almost inexhaustible application possibilities in all areas of life.
Unfortunately, I’m currently not collaborate in an interdisciplinary project.
I hope that these methods will also be used in other courses. Above all, it is important to make students aware of the possibilities.
I am pleased that there is an interdisciplinary platform for exchanging ideas with colleagues and I am excited to see what projects will result from it.
No.
Mr. Nagl, thank you very much for taking the time to answer these questions. We wish you a nice day!
Mr. Nagl, thank you very much for taking the time to answer these questions. We wish you a nice day!