An investigation of machine learning in credit risk
Machine learning algorithms have come to dominate several industries. After decades of resistance from examiners and auditors, machine learning is now moving from the research desk to the application stack for credit scoring and a range of other credit risk applications. This migration is not without new risks and challenges. Much of the research now shifts from how best to create the models to how best to use the models in a regulatory-compliant commercial setting. This article examines the impressive range of machine learning methods and application areas for credit risk. During this investigation, we create a taxonomy to think about how different machine learning components come together to create specific algorithms. Why machine learning succeeds over simple linear methods is explored through a specific loan example. Throughout, we highlight open questions, ideas for improvement, and a framework for thinking about how to choose the best machine learning method for a specific problem.
Sorry, our subscription options are not loading at this time
Please try again later. Contact our customer service if this problem persists.
New to Risk.net? See our subscription options
If you already have an account, please login here.