Bayesian learning models to measure the relative impact of ESG factors on credit ratings

Abstract

AbstractArtificial intelligence methods, based on machine learning models, are rapidly changing financial services, and credit lending in particular, complementing traditional bank lending with platform lending. While financial technologies improve user experience and possibly lower costs, they may increase risks and, in particular, the model risks that derive from inaccurate credit rating assessments. In this paper, we will show how to reduce such model risks, using a S.A.F.E. statistical learning model, which improves: Sustainability, taking environmental, social and governance factors into account; Accuracy, building a model which maximises predictive accuracy; Fairness, merging ESG scores from different data providers, improving their representativeness; Explainability, clarifying the relative contribution of each ESG score to predictive accuracy.