DIAGNOSING THE PATH OF FINANCIAL STABILITY INDEX: A NEW PERSPECTIVE FOR ARTIFICIAL INTELLIGENCE APPLICATIONS

The study highlights the critical challenge of assessing and maintaining financial stability, given its importance for economic growth and also in the face of financial crises. However, it remains vulnerable to internal and external shocks. Traditional methods for assessing financial stability, such as the IMF's Financial Stability Index (FSI) often rely on fixed–weight models that may not adapt well to changing economic conditions. This gap motivates the exploration of Artificial Intelligence applications to enhance the accuracy and predictive power of the FSI. It aims to design an AI-based financial stability index that accurately measures the resilience of the banking sector to crises. The study proposes new algorithm that integrate deep learning models (DLM) outputs to create a fitness function in a proposed genetic algorithm framework. The study used a multi-method AI approach, using machine learning (ML), DLM, and genetic algorithms (GA), to diagnose and correct the FSI Path. GA aims to optimize the component variables of the FSI across various scenarios. The results concluded that AI techniques outperform traditional methods in predicting FSI trends. Also, GA provided scenario-based optimizations, highlighting the FSI's sensitivity to external shocks. The study's contribution lies in its innovative integration of AI into financial stability analysis, providing policymakers with a dynamic tool for crisis response. The study recommends adopting an AI-enhanced financial stability index for early warning systems for shocks and designing macroprudential policies aligned with economic cycles. It also provides a comprehensive framework for developing countries to mitigate financial instability in the face of overlapping crises.