Research & Publications
Academic contributions to machine learning, statistics, and risk management
An Information-Theoretic Framework for Credit Risk Modeling: Unifying Industry Practice with Statistical Theory for Fair and Interpretable Scorecards
arXiv preprint, September 2025
We establish a unified information-theoretic framework revealing Weight of Evidence (WoE), Information Value (IV), and Population Stability Index (PSI) as instances of classical information divergences. Through the delta method applied to WoE transformations, we derive standard errors for IV and PSI, enabling formal hypothesis testing and probabilistic fairness constraints for the first time in credit risk modeling.
Weight of Evidence (WOE), Log Odds, and Standard Errors
xRiskLab Technical Report, 2024
Comprehensive methodology for computing standard errors for Weight of Evidence (WOE) values, with practical implementations.
Industry Insights & Collaborations
Beyond Credit Scoring: A Decision-Making Framework for Profitable Lending
Exploring advanced machine learning techniques and their applications in modern lending decisions, focusing on interpretable AI and regulatory compliance.
University of Edinburgh Business School
External Affiliate at the Credit Research Centre, contributing to research in credit risk management and interpretable machine learning methods for financial applications.
Research Collaboration & Inquiries
Interested in collaborating on research projects or discussing applications of interpretable machine learning in your organization?