FinTwin

This dataset is part of the Financial Digital Twin (FinTwin) project, an open-source research framework designed to generate large-scale, fully synthetic financial market datasets for reproducible Artificial Intelligence (AI), Machine Learning (ML), portfolio optimization, and systemic risk research. The repository produces realistic synthetic financial data by integrating multiple simulation engines, including financial market dynamics, heterogeneous investor behaviors, macroeconomic indicators, financial crisis scenarios, market regime switching, and systemic risk propagation. The generated datasets are intended for academic research, benchmarking, explainable AI, and educational purposes. Because the complete dataset may exceed hundreds of gigabytes or several terabytes, it is distributed as multiple independent chunks. Each chunk follows an identical schema and can be used individually or combined with other chunks to reconstruct the complete dataset.