#USA #SAS #CARY #credit risk #Man Group

New AI-Driven Risk Model Revolutionizes Credit Rating Predictions

In a remarkable advancement for the financial sector, a new credit risk forecasting model has been unveiled, which is said to be among the most accurate ever developed. This innovative approach is the result of a collaborative effort between SAS, a leader in analytics and AI, the investment management firm Man Group plc, the UK-based Pension Insurance Corporation plc, and Stanford University. The aim of this model is significant – it provides early warning signals for predicting changes in corporate credit ratings, allowing investors to gain a competitive edge.

Understanding the Credit Risk Model

Traditional methods of assessing credit ratings often rely on historical data and set guidelines that may not account for rapid market shifts. In contrast, this new model employs advanced machine learning techniques to analyze a multitude of variables, including over 20 years of data from KRIS® (SAS® Kamakura Risk Information Services) that details default probabilities. This robust foundation enables the model to far exceed conventional tools in its ability to predict credit rating changes by categorizing firms into those likely to receive upgrades, downgrades, or remain stable.

As Stas Melnikov, Head of Quantitative Research and Risk Data Solutions at SAS, notes, "This breakthrough approach indicates that investors can achieve significantly better performance than current best practices. By identifying potential credit downgrades or upgrades before they are recognized by institutions and the market, investors can better manage risk, minimize losses, and capitalize on opportunities."

Implications for Portfolio Management

The implications of this model are wide-ranging. By improving the prediction of downgrades, portfolio managers and investors can not only enhance their returns but also reduce their exposure to risk. The model acts as a proactive measure, enabling the identification of potential changes in credit ratings before these events are reflected in the market prices of securities. Consequently, this can protect investors, particularly money managers, banks, and insurance companies, from the fallout of sudden downgrades, which can drastically affect their portfolios and investment strategies.

In current market conditions, where corporate borrowers are grappling with high-interest rates amidst fears of a potential recession, the ability to accurately predict credit downgrades has never been more critical. Investors need to be equipped with timely insights that allow them to react swiftly and strategically to the evolving landscape of risk.

How the Data Works

The model’s architecture draws on a vast dataset encompassing more than half a million observations, providing historical context through variables such as bond spreads, yields, and macroeconomic factors. Among these, the one-year default probability (KDP) has been identified as an essential component, trailing only option-adjusted spread (OAS) and yield-to-maturity (YTM) in its predictive capacity. The insights gained from KDP equip investors with additional data points that can reveal opportunities ahead of the market's response.

Steven Desmyter, President of Man Group, remarked on the effectiveness of this model in a recent LinkedIn post, stating, "The new models significantly outperformed traditional methods, providing better guidance on which companies are at risk of downgrade or upgrade." This validation emphasizes the model’s reliability and its potential to alter industry standards in credit risk forecasting.

A New Era in Credit Rating Management

The aftermath of a credit rating change can greatly impact market perceptions, driving prices up or down as investors adjust their holdings based on newly assigned ratings. In the investment world, where institutional investors often face regulations requiring them to only hold investment-grade debt, the consequences of a downgrade can necessitate immediate divestment, further complicating portfolio management.

SAS, known for its commitment to enhancing risk management across various financial sectors, integrated the Kamakura Corporation into its suite of solutions in 2022. This acquisition was aimed at bolstering its offerings in asset and liability management (ALM) and credit risk management. Today, SAS continues to lead with its innovative solutions that encompass enterprise stress testing, expected credit losses, and risk governance among others.

Conclusion

As we move towards a future shaped by data and AI, the importance of accurate forecasting in credit risk cannot be overstated. This new model represents a significant leap forward in understanding and managing credit risk, offering investors advanced tools to navigate the challenges of today's financial landscape effectively. For more detailed insights and further developments, companies and investors alike are encouraged to engage with SAS to harness the full potential of this groundbreaking technology.