Credit Risk Analysis Using Machine and Deep Learning Models

32 Pages Posted: 17 Apr 2018

See all articles by Peter Martey Addo

Peter Martey Addo

Agence Française de Développement (AFD); Labex ReFi

Dominique Guegan

Université Paris I Panthéon-Sorbonne

Bertrand Hassani

Université Paris I Panthéon-Sorbonne; University College London - Department of Computer Science

Date Written: March 16, 2018

Abstract

Due to the hyper technology associated to Big Data, data availability and computing power, most banks or lending financial institutions are renewing their business models. Credit risk predictions, monitoring, model reliability and effective loan processing are key to decision making and transparency. In this work, we build binary classifiers based on machine and deep learning models on real data in predicting loan default probability. The top 10 important features from these models are selected and then used in the modelling process to test the stability of binary classifiers by comparing performance on separate data. We observe that tree-based models are more stable than models based on multilayer artificial neural networks. This opens several questions relative to the intensive used of deep learning systems in the enterprises.

Keywords: Credit risk, Financial regulation, Data Science, Bigdata, Deep learning

JEL Classification: C55

Suggested Citation

Addo, Peter Martey and Addo, Peter Martey and Guegan, Dominique and Hassani, Bertrand, Credit Risk Analysis Using Machine and Deep Learning Models (March 16, 2018). University Ca' Foscari of Venice, Dept. of Economics Research Paper Series No. 08/WP/2018, Available at SSRN: https://ssrn.com/abstract=3155047 or http://dx.doi.org/10.2139/ssrn.3155047

Peter Martey Addo (Contact Author)

Agence Française de Développement (AFD) ( email )

5, rue Roland Barthes
Paris Cedex 12, 75598
France

Labex ReFi ( email )

79 avenue de la République
Paris, 75011
France

Dominique Guegan

Université Paris I Panthéon-Sorbonne ( email )

106 avenue de lhopital
75634 Paris Cedex 13
Paris, IL
France

Bertrand Hassani

Université Paris I Panthéon-Sorbonne ( email )

17, rue de la Sorbonne
Paris, IL 75005
France

University College London - Department of Computer Science ( email )

United Kingdom

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
613
Abstract Views
1,801
rank
55,333
PlumX Metrics