Critical Phenomena in Financial Markets and Social Networks 

We investigate the interplay of trends and reversion in the markets is analyzed using 13 years of tick data, 30 years of daily data, 330 years of monthly data, and 800 years of annual data. This leads us to a new lattice gas model of financial markets, where the lattice represents the social network of traders, while the gas molecules are shares of a stock. Using advanced tools of statistical physics and conformal field theory, we use the model to explain the empirically observed scaling behavior of financial markets. We work out applications of the new model in asset- and risk management and, in particular, look for early-warning signals of financial market instabilities. 

We expect that our model will lead to a new and deeper understanding and better risk management for financial markets and also for more general social networks. 

Publications:

• C. Schmidhuber, Trends, reversion, and critical phenomena in financial markets. Physica A566 (2021).
• C. Schmidhuber, Financial markets and the phase transition between water and steam. Physica A592 (2022).

Team members:

• Christoph Schmidhuber
• Sara Safari
• Donat Maier

Industry and funding partners:

• SNF Practise-to-Science Grant

Data Driven Financial Risk and Regulatory Reporting

We are developing a new generation of regulatory technology that has the potential to revolutionize the regulatory reporting practice. The development is carried out in close collaboration with industry and regulators with the aim to create solutions that make a difference for the financial industry and society as a whole.

Publications:

• H. Breymann, P. Hauf and C. Künzle. Venturing into New Ways of Regulatory Reporting and Systemic Risk Analysis. Banking Resilience (Chapter 12), 2024.

Team members:

• Prof. Dr. Henriette Elise Breymann
• Prof. Dr. Tim Weingärtner
• Prof. Dr. Walter Farkas
• Dr. Patrick Hauf
• Dr. Francis Parr

Industry and funding partners:

• SNF BRIDGE
• ACTUS Financial Research Foundation
• ECB
• BrickTowers AG

Blockchain Security, Economics and Ledger Models

The project's goal is to advance the current state of the art in blockchain design by designing novel components for blockchains systems, including improvements to consensus, to the identity layers, or data protection layers (using zero-knowledge proofs) which align with people's incentives to use the system. The goal is hereby to design secure components not in isolation, but always bearing in mind that using the system also must follow from rational decisions. After all, blockchain systems are a unique blend of security and economic aspects

As these are hard problems at the forefront of blockchain infrastructure development, the results must undergo peer-review. A critical observation in many of the designs aspects, especially when it comes to implementation, is the underlying ledger state representation, be it UTxO-based, account-based, or object-based. Part of this project is to systematically anticipate the practical impact on deploying certain components in various ledger models.

Publications:

• C. Badertscher, M. Campanelli, M. Ciampi, L. Russo, and L. Siniscalchi. Universally Composable SNARKs with Transparent Setup without Programmable Random Oracle. International Cryptology Conference (Crypto) 2025.
• C. Badertscher, F. Banfi and J. Diaz Vico. What Did Come Out of it? Analysis and Improvements of DIDComm Messaging. 2024 ACM SIGSAC Conference on Computer and Communications Security.

Team members:

• Christian Badertscher
• Michael Läderach

Industry and funding partners:

• This is a directly financed project with an industry partner.

Forecasting German GDP

Forecasting German GDP is a collaborative project involving our team as well as IWH (Leibniz-Institut für Wirtschaftsforschung Halle) and CIRET (Montreal). Specifically, our group has developed a novel prediction algorithm that emphasizes mid- to long-term forecast horizons, extending beyond two quarters. While traditional forecast optimization criteria often focus on a single objective—such as minimizing mean squared error (MSE)—we introduce a new Smooth Sign Accuracy (SSA) approach that concurrently evaluates sign accuracy, MSE, and the frequency of zero-crossings in predictions. Zero-crossings, which represent sign changes in the stationary growth rate of a non-stationary time series, serve as crucial indicators of transitions between expansion and contraction phases, playing a vital role in decision-making and control processes. The SSA framework effectively captures and balances these diverse aspects by explicitly recognizing the inherent trade-off—termed the accuracy-smoothness (AS) dilemma—faced by predictors. 

Team members:

• Marc Wildi

Additional Links:

• Github-Page of the Project

Commodities and Alternative Investment Strategies

A common theme across the different research projects is empirical analysis of (private or public) financial data, where the work typically explores new or alternative methods with respect to established industry standards. This require both methodology and data modelling expertise as well as industry experience with financial data in the given context, to produce reliable and actionable results.

Within the area of Commodity and Alternative Investments, the research often concerns different market factors that are central to the given investment or trading thesis. For example, this could be a quantitative study of a fundamental factor associated with supply and demand expected to impact the market of a commodity, or a technical factor used in the hypothesis behind a systematic risk-premia/QIS strategy. Another common subject is risk within this specific context. This could relate to individual markets, such as exploring a hedging approach or a quantitative risk metric, as well as portfolio risk where the circumstances related to a commodity or alternative investment portfolio may align better with a customized approach.

Team members:

• Stefan Ask

“Great things in business are never done by one person, they are done by a team of people” ― Steve Jobs.