Credit Risk Analysis Using Machine and Deep Learning Models
1
UP1 -
Université Paris 1 Panthéon-Sorbonne
2 CES - Centre d'économie de la Sorbonne
3 Labex ReFi
4 IPAG Business School
5 University of Ca’ Foscari [Venice, Italy]
6 AFD - Agence française de développement
7 Capgemini Consulting [Paris]
8 UCL-CS - Department of Computer science [University College of London]
2 CES - Centre d'économie de la Sorbonne
3 Labex ReFi
4 IPAG Business School
5 University of Ca’ Foscari [Venice, Italy]
6 AFD - Agence française de développement
7 Capgemini Consulting [Paris]
8 UCL-CS - Department of Computer science [University College of London]
Dominique Guegan
- Fonction : Auteur
- PersonId : 18037
- IdHAL : dominique-guegan
- ORCID : 0000-0003-4214-1429
- IdRef : 026905809
Peter Martey Addo
- Fonction : Auteur
- PersonId : 963521
- ORCID : 0000-0001-5895-9108
- IdRef : 175914710
Bertrand Hassani
- Fonction : Auteur
- PersonId : 937147
- IdRef : 158318099
Résumé
Due to the advanced technology associated with Big Data, data availability and computing power, most banks or lending 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 modeling process to test the stability of binary classifiers by comparing their performance on separate data. We observe that the tree-based models are more stable than the models based on multilayer artificial neural networks. This opens several questions relative to the intensive use of deep learning systems in enterprises.
Format du dépôt | Fichier |
---|---|
Type de dépôt | Article dans une revue |
Titre |
en
Credit Risk Analysis Using Machine and Deep Learning Models
|
Résumé |
en
Due to the advanced technology associated with Big Data, data availability and computing power, most banks or lending 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 modeling process to test the stability of binary classifiers by comparing their performance on separate data. We observe that the tree-based models are more stable than the models based on multilayer artificial neural networks. This opens several questions relative to the intensive use of deep learning systems in enterprises.
|
Auteur(s) |
Dominique Guegan
1, 2, 3, 4, 5
, Peter Martey Addo
6, 3
, Bertrand Hassani
3, 2, 7, 8
1
UP1 -
Université Paris 1 Panthéon-Sorbonne
( 7550 )
- 12 place du Panthéon, 75231 Paris Cedex 05
- France
2
CES -
Centre d'économie de la Sorbonne
( 15080 )
- Maison des Sciences Économiques - 106-112 Boulevard de l'Hôpital - 75647 Paris Cedex 13
- France
3
Labex ReFi
( 472714 )
- France
4
IPAG Business School
( 542840 )
- 184 boulevard Saint-Germain, 75006 Paris
- France
5
University of Ca’ Foscari [Venice, Italy]
( 461222 )
- Dorsoduro, 3246, 30123 Venice, Italy
Universita di Venezia - Ca' Foscari
- Italie
6
AFD -
Agence française de développement
( 133737 )
- 5 Rue Roland Barthes - 75598 PARIS CEDEX 12
- France
7
Capgemini Consulting [Paris]
( 458132 )
- Tour Europlaza - 20, avenue André Prothin - 92927 Paris - La Défense Cedex
- France
8
UCL-CS -
Department of Computer science [University College of London]
( 84375 )
- University College London Computer science department Adastral Park - Martlesham IP5 3RE - UK
- Royaume-Uni
|
Langue du document |
Anglais
|
Nom de la revue |
|
Vulgarisation |
Non
|
Comité de lecture |
Oui
|
Audience |
Internationale
|
Date de publication |
2018
|
Date de publication électronique |
2018-04-16
|
Titre de la collection |
Computational Methods for Risk Management in Economics and Finance
|
Volume |
6
|
Numéro |
2
|
Page/Identifiant |
38
|
URL éditeur |
http://www.mdpi.com/2227-9091/6/2/38
|
Domaine(s) |
|
Mots-clés (JEL) |
|
Projet(s) ANR |
|
Mots-clés |
en
financial regulation, deep learning, Big data, data science, credit risk
|
DOI | 10.3390/risks6020038 |
Origine :
Fichiers éditeurs autorisés sur une archive ouverte
Loading...