Background: Large fluctuations, instabilities, trends and uncertainty of financial markets constitute a substantial challenge for asset management companies, pension funds and regulators. Nowadays, most asset management companies and financial institutions follow a so-called systematic trading approach in their investment decisions. Systematic trading refers to applying predefined, rule-based trading strategies for buy- and sell orders.

However, automated or rules-based trading activities bring certain risks for market participants and the whole financial market. In times of increased market volatility, market turmoil or so-called market sell-offs, investors applying similar trading rules might undertake the same actions, escalating and increasing systemic market risk through such behavior. Such situations have been frequently observed on financial markets for instance, in March 2020 (sell-off related to the Covid pandemic), during the European Sovereign Debt crisis and the global financial crisis 2007-08.

Rationale: Despite advancements in econometric methods, the detection of asset price bubbles and structural breaks remains uncertain and underexplored, particularly in the context of classic financial assets. Recent research underscores the powerful role of narratives in financial decision-making, suggesting that integrating textual analysis could enhance market understanding and prediction accuracy.

Overall objectives: The project aims to develop a comprehensive framework that utilizes advanced machine learning and NLP techniques to predict market outcomes, detect asset price bubbles, and identify structural breaks using diverse data sources, including financial data and narrative content from text, speech, and multimedia.

Specific aims: The project has three main goals: first, to validate and refine existing econometric models using real-world financial data; second, to integrate narrative analysis to understand and predict market behaviors and asset price dynamics; and third, to create a multidimensional AI and ML framework that enhances the detection of market anomalies and forecasts financial trends.

Methods: The approach involves collecting and processing a wide range of data, including stock prices, macroeconomic indicators, and textual content from the web. We will employ text mining and NLP techniques to analyze sentiment, narrative structures, and their impact on market movements. Developing and testing new AI models that combine traditional financial analysis with narrative insights to predict market changes and detect structural breaks will also be a key methodological focus.

Expected results: The project is expected to yield enhanced models for predicting financial markets with greater accuracy, especially in detecting asset price bubbles and structural breaks. It will also produce novel datasets and measurement techniques that provide deeper insights into the interplay between market narratives and financial indicators. Practical tools for asset management and regulatory bodies to better anticipate and react to market crises will be developed.

Impact for the field: This research could significantly transform how financial markets are understood and modeled. It aims to bridge the gap between traditional econometric approaches and modern AI techniques, providing new insights into the role of narratives and quantitative data in financial decision-making. The results are anticipated to offer valuable tools for risk management, asset pricing, and optimizing trading strategies, potentially influencing policy and regulatory practices globally. This approach is particularly relevant in understanding the complexities of today's non-linear, dynamic financial markets, enhancing the robustness of financial models against future economic crises.