Forecasting nonperforming loans using machine learning

Abdullah, Mohammad; Chowdhury, Mohammad Ashraful Ferdous; Uddin, Ajim; Moudud‐Ul‐Huq, Syed

doi:10.1002/for.2977

Cited by 9 publications

(2 citation statements)

References 68 publications

(113 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their experimental results demonstrated a significant improvement, surpassing the 80% accuracy threshold, offering promising implications for the financial sector. In (Abdullah et al, 2023), the research objective focused on forecasting nonperforming loans in financial institutions, recognizing their significance in overall performance. Employing machine learning techniques, the study utilized quarterly cross-sectional data from 322 banks in 15 emerging countries.…”

Section: Related Workmentioning

confidence: 99%

An Intelligent Model to Assess the Credit Risk in Egyptian Banks

Fathy,

Marie,

Yehia

2024

المجلة العلمية للبحوث والدراسات التجارية

View full text Add to dashboard Cite

In the realm of financial and banking institutions, the art of forecasting and assessing banking risks holds paramount significance. Preserving the financial stability of banks is contingent upon adept risk management, a cornerstone in enhancing overall bank performance. Moreover, the effectiveness of financial and banking institutions can be gauged by their ability to systematically evaluate and mitigate risks. Among these risks, the assessment of banking credit risks looms large in contemporary times, given the heightened necessity for decision-makers to anticipate the likelihood of loan defaults. However, one formidable challenge persists: the inadequate assessment of banking credit risks. This challenge stems from the multifaceted factors that influence risk assessment and the soundness of credit decisions. In response to this pressing issue, our research presents a predictive model employing machine learning (ML) algorithms. Our objective is to facilitate informed credit decision-making and safeguard the financial assets of banks. In pursuit of this aim, we employed five machine learning classification algorithms: Artificial Neural Networks (ANN), Random Forest (RF), Logistic Regression (LR), Decision Tree (DT) and XGBoost (XGB). To ensure the robustness of our study, we utilized a real-world dataset gleaned from the historical records of a prominent Egyptian bank. Subsequently, we assessed the performance of our model based on key metrics such as accuracy, precision, recall, and the f1 score. The results showed that XGB exhibited the highest accuracy, underlining the potential for ML algorithms to revolutionize the assessment of banking credit risks.

show abstract

Section: Related Workmentioning

confidence: 99%

An Intelligent Model to Assess the Credit Risk in Egyptian Banks

Fathy,

Marie,

Yehia

2024

المجلة العلمية للبحوث والدراسات التجارية

View full text Add to dashboard Cite

show abstract

“…Abdullah and colleagues (2023) [10] explored a range of machine learning techniques to forecast nonperforming loans within financial institutions in emerging countries. By examining the data from 322 banks spanning 15 nations, their comprehensive analysis revealed that advanced machine learning models, particularly random forest, surpassed linear methods, achieving an accuracy of 76.10%.…”

Section: Related Researchmentioning

confidence: 99%

Machine Learning for an Enhanced Credit Risk Analysis: A Comparative Study of Loan Approval Prediction Models Integrating Mental Health Data

Alagic,

Zivic,

Kadusic

et al. 2024

MAKE

View full text Add to dashboard Cite

The number of loan requests is rapidly growing worldwide representing a multi-billion-dollar business in the credit approval industry. Large data volumes extracted from the banking transactions that represent customers’ behavior are available, but processing loan applications is a complex and time-consuming task for banking institutions. In 2022, over 20 million Americans had open loans, totaling USD 178 billion in debt, although over 20% of loan applications were rejected. Numerous statistical methods have been deployed to estimate loan risks opening the field to estimate whether machine learning techniques can better predict the potential risks. To study the machine learning paradigm in this sector, the mental health dataset and loan approval dataset presenting survey results from 1991 individuals are used as inputs to experiment with the credit risk prediction ability of the chosen machine learning algorithms. Giving a comprehensive comparative analysis, this paper shows how the chosen machine learning algorithms can distinguish between normal and risky loan customers who might never pay their debts back. The results from the tested algorithms show that XGBoost achieves the highest accuracy of 84% in the first dataset, surpassing gradient boost (83%) and KNN (83%). In the second dataset, random forest achieved the highest accuracy of 85%, followed by decision tree and KNN with 83%. Alongside accuracy, the precision, recall, and overall performance of the algorithms were tested and a confusion matrix analysis was performed producing numerical results that emphasized the superior performance of XGBoost and random forest in the classification tasks in the first dataset, and XGBoost and decision tree in the second dataset. Researchers and practitioners can rely on these findings to form their model selection process and enhance the accuracy and precision of their classification models.

show abstract