Prof. Dr. Maxim Ulrich – Karlsruhe Institute of Technology (KIT)

Since 2014, Prof. Dr. Maxim Ulrich has held the Chair of Financial Economics and Risk Management at KIT. Before joining KIT, he served from 2008 - 2013 as a Tenure-Track Professor at Columbia Business School in New York, where he taught investment and asset pricing at the MBA, Executive MBA, Master, and Ph.D. levels. His research has been presented at leading academic conferences—including the AFA, WFA, EFA, EEA, and NBER Asset Pricing Meetings—and published in journals such as the Journal of Monetary Economics , the Review of Financial Studies and the Review of Derivatives Research .

The Chair's research is grounded in the idea that predictive information in financial markets is scarce, noisy, and often overlooked by conventional models. Our work focuses on separating signal from noise in complex financial systems—particularly in real-time, high-frequency environments—by combining economic reasoning with machine learning, statistical modeling, and large-scale computation.

We specialize in extracting forward-looking signals from financial markets—especially from high-frequency option data. These signals are used to estimate return densities, crash risk indicators, and expected returns, often at millisecond resolution. This research forms the basis for both academic inquiry and practical applications in financial data science and machine learning.

Research at the Chair is structured into two interdisciplinary groups:

This group develops and adapts machine learning and AI methods for financial and business data. Key challenges include noisy signals, limited training data, non-stationarity, and the demand for interpretability. Recent projects involve mixture-density neural networks for forecasting equity return distributions, and reinforcement learning systems for robust portfolio decision-making. The AI-Finance group collaborates closely with computer scientists, statisticians, and financial economists within and beyond KIT.

The C-RAM group focuses on empirical asset pricing, financial risk modeling, and high-performance analytics using economic and market data. A key asset of the group is a proprietary, millisecond-level dataset of predictive signals extracted from international option markets—constructed over several years as part of an applied innovation initiative. While its initial goal included entrepreneurial applications, the resulting dataset now forms a unique research and teaching infrastructure used extensively across thesis supervision, KABFI doctoral work, and advanced classroom projects.

Using this data, C-RAM researchers study forward-looking risk premium, tail risk, and return densities under real-time market conditions—topics that are of growing importance in financial data science and machine learning. Ongoing projects also include model-free estimation of dividend risk premium and the analysis of central bank communication using natural language processing.

Both groups are tightly integrated with the KABFI Ph.D. Research Training Group on Artificial Intelligence for Business and Financial Market Investigations, and collaborate closely with the Graduate School Computational and Data Science (KCDS) at KIT. We are also actively exploring partnerships with Ph.D. programs in computer science at other German universities to co-train doctoral researchers working at the intersection of finance, machine learning, and data science.

Our teaching is directly informed by our research. Core modules such as the Bachelor-level Financial Data Science course and the Master-level Advanced Machine Learning & Data Science module engage students in real-time data analysis and applied modeling. Students regularly work with the same datasets, methods, and tools used in current research.

Alumni of the Chair have gone on to competitive roles in quantitative finance, data science, and research across global hubs such as New York, London, Zurich, and Frankfurt. As one graduate now at Morgan Stanley, New York, noted:
"Your classes truly prepared me for the interviews, and our project (master thesis) gave me a lot to discuss. Topics ranging from factor models to stochastic discount factor and machine learning were part of the conversations, and the way you taught us about these subjects gave me the confidence to speak about them."

Another alumnus, with a first job as quant researcher in asset management at Barclays in London, emphasized:
“The interviews were like your lecture notes… The ability to connect math with economic intuition made the difference.”

We welcome with researchers and institutions across collaborations and encourage ambitious students to engage early through research projects, thesis work, and doctoral training.