Performance Assessment of Logistic Regression (LR), Artificial Neural Network (ANN), Fuzzy Inference System (FIS) and Adaptive Neuro-Fuzzy System (ANFIS) in Predicting Default Probability: The Case of a Tunisian Islamic Bank

Nadia Ayed () and Khemaies Bougatef ()
Additional contact information
Nadia Ayed: Higher Institute of Computer Science and Management of Kairouan, University of Kairouan
Khemaies Bougatef: Higher Institute of Computer Science and Management of Kairouan, University of Kairouan

Computational Economics, 2024, vol. 64, issue 3, No 17, 1803-1835

Abstract: Abstract This paper aims to compare the performance of four credit scoring models, namely logistic regression (LR), artificial neural network (ANN), fuzzy inference system (FIS) and adaptive neuro-fuzzy inference system (ANFIS) in predicting default probability. We use a sample of 1045 consumer credits (after oversampling the initial sample of 660 customers) granted by a Tunisian Islamic bank. The six explanatory variables retained to predict the probability of default are: residual wage, age, job tenure, profession, financing type and region of residence. Our findings reveal that ANFIS and LR have the highest discriminating power (AUC = 0.9). Regarding the type I error (false-positive) and the type II (false-negative) error, it has been proved that ANFIS has the lowest misclassification costs (MC = 0.15). The outperformance of the ANFIS comes from combining the advantages of neural networks with a fuzzy inference system. Thus, our results suggest that the ANFIS seems to be the most efficient and transparent technique for predicting credit risk in Islamic banks. Unlike ANN, the ANFIS allows bankers to justify the reasons behind the rejection of credit applications.

Keywords: Credit scoring; Islamic banking; Oversampling; Logistic regression; Neural network; Fuzzy inference system; Adaptive neuro-fuzzy inference system (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s10614-023-10496-y Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:kap:compec:v:64:y:2024:i:3:d:10.1007_s10614-023-10496-y

Ordering information: This journal article can be ordered from
http://www.springer. ... ry/journal/10614/PS2

DOI: 10.1007/s10614-023-10496-y

Access Statistics for this article

Computational Economics is currently edited by Hans Amman

More articles in Computational Economics from Springer, Society for Computational Economics Contact information at EDIRC.
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().