Machine Learning in Credit Risk Modeling: Empirical Application of Neural Network Approaches

Abstract

Motivated by massive development and the ongoing practice of machine learning (ML) approaches in credit risk modeling, this chapter addresses theoretical aspects of machine learning and credit default prediction. This chapter also discusses the properties of mostly used and robust machine learning approaches in credit default prediction. The objective of the chapter does not do empirical analysis, however, to show the practical application of ML approaches in credit default prediction, to the end, this chapter presents an empirical example of trendy classifier neural network approaches on real-world credit datasets.