The Credit Research Initiative (CRI) at the National University of Singapore today announced the launch of two significant resources aimed at transforming credit risk assessment for Micro, Small, and Medium Enterprises (MSMEs): the TRACE framework whitepaper and a corresponding pilot dataset for Singapore’s Food & Beverage (F&B) industry.
Building on its globally recognized research in forecasting the Probability of Default (PD) for public companies, CRI is now addressing the complex and often opaque world of private MSME credit risk. This new initiative provides financial institutions, researchers, and policymakers with powerful tools to better understand and manage risk in this vital economic sector.
The culmination of this extensive research is the TRACE (An Interpretable, High-Performance MSME Credit Risk Assessment) framework. It offers a scientifically rigorous and practical approach to assessing private MSME credit risk, powered by advanced AI and machine learning. Crucially, the framework is designed to maintain full transparency, with key features including Intelligent Data Integration, ‘Glass Box’ Modelling, and a structure aligned with regulatory expectations.
To demonstrate the framework's capabilities and data coverage, CRI has released a pilot dataset focusing on Singapore’s F&B industry. The dataset presents companies in numerically-labeled groups based on their calculated Probability of Default (PD) range.
Availability and Further Engagement
The full TRACE framework whitepaper and the pilot dataset are now available for public access on the CRI website at: https://nuscri.org/en/private_company/
CRI invites industry partners, financial institutions, and researchers to explore potential collaborations, including the development of custom datasets for other sectors or the application of the TRACE framework to proprietary data. The team also welcomes any questions, comments, or suggestions.