Reject inference methods in credit scoring
Type de document :
Article dans une revue scientifique: Article original
PMID :
URL permanente :
Titre :
Reject inference methods in credit scoring
Auteur(s) :
Ehrhardt, A. [Auteur]
Biernacki, Christophe [Auteur]
Laboratoire Paul Painlevé - UMR 8524 [LPP]
Vandewalle, Vincent [Auteur]
METRICS : Evaluation des technologies de santé et des pratiques médicales - ULR 2694
Heinrich, Philippe [Auteur]
Laboratoire Paul Painlevé - UMR 8524 [LPP]
Beben, S. [Auteur]
Biernacki, Christophe [Auteur]
Laboratoire Paul Painlevé - UMR 8524 [LPP]
Vandewalle, Vincent [Auteur]
METRICS : Evaluation des technologies de santé et des pratiques médicales - ULR 2694
Heinrich, Philippe [Auteur]
Laboratoire Paul Painlevé - UMR 8524 [LPP]
Beben, S. [Auteur]
Titre de la revue :
Journal of Applied Statistics
Nom court de la revue :
J. Appl. Stat.
Numéro :
48
Pagination :
p. 2734-2754
Date de publication :
2021
ISSN :
0266-4763
Mot(s)-clé(s) en anglais :
semi-supervised learning
scorecard
data augmentation
scoring
credit risk
Reject inference
scorecard
data augmentation
scoring
credit risk
Reject inference
Discipline(s) HAL :
Sciences du Vivant [q-bio]
Résumé en anglais : [en]
The granting process is based on the probability that the applicant will refund his/her loan given his/her characteristics. This probability, also called score, is learnt based on a dataset in which rejected applicants are ...
Lire la suite >The granting process is based on the probability that the applicant will refund his/her loan given his/her characteristics. This probability, also called score, is learnt based on a dataset in which rejected applicants are excluded. Thus, the population on which the score is used is different from the learning population. Many “reject inference” methods try to exploit the data available from the rejected applicants in the learning process. However, most of these methods are empirical and lack of formalization of their assumptions, and of their expected theoretical properties. We formalize such hidden assumptions in a general missing data setting for some of the most common reject inference methods. It reveals that hidden modelling is mostly incomplete, thus prohibiting to compare existing methods within the general model selection mechanism (except by financing “non-fundable” applicants). So, we assess performance of the methods on both simulated data and real data (from CACF, a major European loan issuer). Unsurprisingly, no method seems uniformly dominant. Both these theoretical and empirical results not only reinforce the idea to carefully use the classical reject inference methods but also to invest in future research works for designing model-based reject inference methods (without financing “non-fundable” applicants).Lire moins >
Lire la suite >The granting process is based on the probability that the applicant will refund his/her loan given his/her characteristics. This probability, also called score, is learnt based on a dataset in which rejected applicants are excluded. Thus, the population on which the score is used is different from the learning population. Many “reject inference” methods try to exploit the data available from the rejected applicants in the learning process. However, most of these methods are empirical and lack of formalization of their assumptions, and of their expected theoretical properties. We formalize such hidden assumptions in a general missing data setting for some of the most common reject inference methods. It reveals that hidden modelling is mostly incomplete, thus prohibiting to compare existing methods within the general model selection mechanism (except by financing “non-fundable” applicants). So, we assess performance of the methods on both simulated data and real data (from CACF, a major European loan issuer). Unsurprisingly, no method seems uniformly dominant. Both these theoretical and empirical results not only reinforce the idea to carefully use the classical reject inference methods but also to invest in future research works for designing model-based reject inference methods (without financing “non-fundable” applicants).Lire moins >
Langue :
Anglais
Audience :
Internationale
Vulgarisation :
Non
Établissement(s) :
Université de Lille
CHU Lille
CHU Lille
Collections :
Date de dépôt :
2023-11-15T03:52:42Z
2024-04-08T14:37:17Z
2024-04-08T14:37:17Z
Fichiers
- document
- Accès libre
- Accéder au document