This paper examines the regulatory impact on the European Banking Sector using advanced deep learning techniques to analyze the relationship between Sustainable Finance guidelines and the SX7P Index from January 2012 to December 2023. Utilizing Long Short-Term Memory Auto-encoder (LSTM-AE), Variational Autoencoder (VAE), and Convolutional Neural Network (CNN) for anomaly detection, the study compares anomalies and investigates their correlation with European Banking Authority (EBA) events and Sustainable Finance guidelines from January 2020 to December 2023. Through the analysis of 43 pertinent EBA documents, the research identifies patterns and variations in anomalies, assessing their association with regulatory changes. The results reveal significant anomalies aligning with regulatory events, indicating a potential causal relationship. Notably, the VAE methodology shows the strongest correlation between EBA Sustainable Finance events and anomalies. This research advances the understanding of deep learning applications in financial markets and offers valuable insights for policymakers and financial institutions regarding regulatory shifts in Sustainable Finance.

This study investigates the predictive power of transition risk for the European stock market by incorporating a text-based Transition Risk Index into an XGBoost forecasting model for the European Index STOXX600. Utilizing daily data from January 2005 to December 2023, we integrate standard macroeconomic factors—such as exchange rates, gold prices, and interest rates—with an NLP-derived Transition Risk Index that captures shifts in climate policies, technological innovations, and investor sentiment. Through an extensive feature engineering process and rigorous hyperparameter tuning (via cross-validation), we assess the relative contribution of each predictor using both feature importance rankings and SHAP (SHapley Additive exPlanations) analysis. Our findings reveal that, while conventional macro-financial variables remain the dominant drivers of STOXX600 price dynamics, transition risk exerts only a modest influence on short-term market forecasts. This suggests that near-term valuation processes may not yet fully integrate sustainability considerations. However, as climate policies evolve and investor awareness grows, the transition risk’s role may become more pronounced over longer horizons. The results underscore both the potential and challenges in quantifying transition risk, offering a robust perspective on incorporating sustainability-driven metrics into real-time market analytics and predictive models. Furthermore, this research highlights the importance of refining transition risk measurement methodologies, as existing models may not fully capture the speed and complexity of regulatory shifts, technological advancements, and investor sentiment changes.

Financial markets do not evolve in a stable or uniform way. They often move through different phases, shaped by changes in risk, returns, volatility, and investor behaviour. These shifts are especially important in emerging markets, where lower liquidity, exposure to global shocks, and domestic vulnerabilities can make transitions from calm periods to stress episodes faster and more severe. This paper examines whether such regimes can be identified and forecast in the Romanian equity market, using the BET index as the main reference. The analysis covers the period from November 2017 to November 2025. In the first stage, a three-state Gaussian Hidden Markov Model is used to identify latent market regimes. The estimated states are then interpreted as calm, stress, and panic regimes based on their conditional realised volatility. In the second stage, the regimes are forecast one day ahead using an XGBoost classifier trained on macro-financial variables, including global risk indicators, cross-asset returns, sovereign yield curve measures, commodity prices, and domestic macroeconomic indicators. The results show strong overall predictive performance. Calm periods are identified with high reliability, while stress regimes are detected with meaningful sensitivity. Panic regimes, however, remain harder to classify because they occur less frequently, which highlights the difficulty of forecasting extreme market conditions in emerging economies.

This study examines whether the Romanian equity market exhibits persistent risk regimes, whether these regimes can be forecast at short horizons, and whether such forecasts can improve portfolio allocation. The analysis uses daily BET index data over 2016–2025 and applies a two-step empirical framework. A Hidden Markov Model is first used to identify latent market states based on returns and realised volatility, followed by an XGBoost classifier that predicts regimes one day ahead using market, volatility, domestic macro-financial, and global indicators. The results reveal three distinct regimes, interpreted as calm, stress, and panic, each reflecting different levels of market risk. The forecasting model performs well in the calm and stress regimes, while predictions for the panic state are less precise but remain informative from a risk-management perspective. Most classification errors occur between neighbouring regimes rather than between extreme states. The economic evaluation shows that a regime-probability portfolio strategy with volatility targeting outperforms a passive Buy-and-Hold investment in the BET index, generating higher cumulative performance and lower drawdowns. Overall, the findings indicate that risk regimes in the Romanian equity market are persistent, short-term predictable, and economically useful for systematic portfolio allocation.