Review on the Innovation of Investment Banks’ Credit Risk Assessment System in a Highly Volatile Market

In highly volatile markets, traditional credit risk assessment systems for investment banks face critical limitations, including reliance on outdated data, linear assumptions, and inadequate integration of non-financial factors. This study proposes a machine learning-driven framework to address these gaps, leveraging real-time multi-source data (e.g., macroeconomic indicators, market sentiment, transactional behavior) and nonlinear algorithms to enhance predictive accuracy. Case analysis, including JPMorgan Chase’s post-2008 reforms, validates the system’s effectiveness in mitigating risks during extreme market fluctuations. Results highlight significant improvements in real-time risk identification and adaptability to dynamic environments. Challenges such as model interpretability and data quality persist, necessitating future research on explainable AI and ESG integration. The findings provide actionable insights for modernizing risk management practices, offering a robust pathway to bolster financial stability in increasingly unpredictable markets.