The research team from the University of Liechtenstein used the US regional banking crisis of 2023 as a case study, which served as an example of how machine learning can be used effectively to identify potential risks at an early stage. Using modern AI models, the scientists analyzed a variety of macroeconomic and bank-specific data to identify warning signals that indicate impending market distortions.
The practical relevance of the results was particularly impressive: The model is not only able to identify potential crises, but can also derive specific recommendations for action for investors. For example, it was shown that an investor who follows the model's signals can exit the market at an early stage in the event of impending negative market developments. This not only leads to a reduction in the risk of loss, but also significantly improves the portfolio's key figures - in particular the final assets and the Sharpe ratio.
The results presented underline the growing importance of data-driven methods in finance. They impressively demonstrate how scientific research can contribute to making the financial market more robust and crisis-resistant. The project is an example of the successful combination of theory and practice and a further indication of the innovative research at the University of Liechtenstein in the field of financial economics and AI.