Mathematical Finance & Financial Data Science Seminar

A Sparsity Algorithm for Finding Optimal Counterfactual Explanations: Application to Corporate Credit Rating

Speaker: Dan Wang, Moody’s Analytics Inc.

Location: Online Zoom access provided to registrants

Date: Tuesday, November 22, 2022, 5:30 p.m.


Machine learning methods used in finance for corporate credit rating lack transparency as to which accounting features are important for the respective rating. A counterfactual explanation is a methodology that attempts to find the smallest modification of the input values which changes the prediction of a learned algorithm to a new output, other than the original one. In this work, we reformulate the problem of finding a counterfactual explanation as an optimization problem. We propose a “sparsity algorithm” which solves this problem. We validate the novel algorithm with synthetically generated data and we apply it to quarterly financial statements from companies in the US market. We provide evidence that the counterfactual explanation can capture the majority of features that change between two quarters when corporate ratings improve. The results obtained show that the higher the rating of a company, the greater the ``effort'' required to further improve credit rating.

Speaker Bio:

Dan is currently director of Machine Learning at Moody's Analytics in New York, where he focuses on natural language processing, recommendation system, deep learning and quantitative methods on RISK (credit, compliance, reputation risk and KYC). He graduated with a PhD from Stevens Institute of Technology Financial Engineering department. His PhD research focused on machine learning applications in credit risk.


This event is free, but requires registration.  Please click here to register.  You will then receive the Zoom link by email about a day or so before the event. 

Slides for Dan Wang's Talk