Statistical and machine learning models in credit scoring: A systematic literature survey

• Published in: Applied soft computing Vol. 91; p. 106263
• Main Authors: Dastile, Xolani, Celik, Turgay, Potsane, Moshe
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.06.2020
• Subjects: Credit scoring; Deep learning; Machine learning; Statistical learning; Systematic literature survey; Deep learning; Statistical learning; Systematic literature survey; Credit scoring; Machine learning
• ISSN: 1568-4946; 1872-9681
• DOI: 10.1016/j.asoc.2020.106263

In practice, as a well-known statistical method, the logistic regression model is used to evaluate the credit-worthiness of borrowers due to its simplicity and transparency in predictions. However, in literature, sophisticated machine learning models can be found that can replace the logistic regression model. Despite the advances and applications of machine learning models in credit scoring, there are still two major issues: the incapability of some of the machine learning models to explain predictions; and the issue of imbalanced datasets. As such, there is a need for a thorough survey of recent literature in credit scoring. This article employs a systematic literature survey approach to systematically review statistical and machine learning models in credit scoring, to identify limitations in literature, to propose a guiding machine learning framework, and to point to emerging directions. This literature survey is based on 74 primary studies, such as journal and conference articles, that were published between 2010 and 2018. According to the meta-analysis of this literature survey, we found that in general, an ensemble of classifiers performs better than single classifiers. Although deep learning models have not been applied extensively in credit scoring literature, they show promising results. •Ensemble classifiers outperform single classifiers.•Deep Learning models to replace classical models in credit scoring.•Convolutional Neural Nets outperform statistical and classical ML models.•Non-transparency of DL models serves as a bottleneck for them to be used.