Determinants of non-performing loans: An explainable ensemble and deep neural network approach

Nwafor, Chioma; Nwafor, Obumneme

doi:10.1016/j.frl.2023.104084

Cited by 6 publications

(4 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, Thakor (2014) argued that financial institutions with a large amount of equity capital tend provide greater loans, generate greater liquidity, and demonstrate greater resilience during economic downturns. This findings, along with the research conducted by Ijaz et al (2020), Khaled et al (2021), and Nwafor & Nwafor (2023), highlight the impact of non-performing loans (NPL) on the occurence of banking crises. They suggest that if not maintained properly, it might lead to a deterioration in banks' profitability and ultimately threaten stability due to their negative impact on the real economy.…”

Section: Resultssupporting

confidence: 56%

“…Conversely, Šeho et al (2023) discovered a negative correlation between loan and financing diversity and Islamic bank stability. Impaired Loan (IL) has a significant impact on bank stability, as indicated by Ijaz et al (2020), Khaled et al (2021), and Nwafor & Nwafor (2023). Liquid Assets (LA) has a significant relationship to stability (Kurtoglu & Durusu-Ciftci, 2024).…”

Section: Methodsmentioning

confidence: 99%

“…In the studies of Ali & Puah (2019) and Cihak & Hesse (2010), banks have shown the impact of factors such as bank size on bank stability. Then, the studies conducted by Ijaz et al (2020), Khaled et al (2021), andNwafor &Nwafor (2023) have all affirmed the significant impact of non-performing loans on bank stability. However, diversification's effect on bank stability varies across studies.…”

Section: Introductionmentioning

confidence: 93%

See 2 more Smart Citations

Islamic and Conventional Banks Stability in Indonesia

Khan,

Aziza,

Nawaz

2024

Etikonomi

View full text Add to dashboard Cite

Research Originality: The stability of banking institutions is crucial for the overall financial health of a country. The originality of this paper lies in its innovative use of quantile regression analysis to provide a nuanced understanding of the factors influencing bank stability, particularly in the Indonesian context.Research Objectives: This research identifies factors influencing bank stability in Islamic and conventional banks in Indonesia.Research Methods: Utilizing quantile regression analysis to measure the impact of various factors at different quantiles, addressing the existing mixed findings in the literature. This study utilizes 15 years of data, encompassing 61 observations for Islamic banks and 779 observations for conventional banks in Indonesia.Empirical Results: The study identifies impaired loans as the primary determinant of bank stability in Indonesia, with a notably greater impact on conventional banks compared to Islamic banks. Notably, the upper quantile of Islamic banks indicates a decline or a lesser impact on bank stability. The paper additionally concludes that the stability of Islamic banks in Indonesia is comparatively lower than conventional banks.Implications: By offering insights into the different levels of factors affecting bank stability, the study contributes to the existing literature, sheds light on unique dynamics, and highlights potential recommendations for enhancing bank stability, especially in the context of non-performing loans.JEL Classification: C21, C23, G21How to Cite:Khan, S., Aziza, N. A., & Nawaz, A. D. (2024). Islamic and Conventional Banks Stability in Indonesia. Etikonomi, 23(1), 167 – 182. https://doi.org/10.15408/etk.v23i1.33879.

show abstract

Section: Resultssupporting

confidence: 56%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 93%

See 1 more Smart Citation

Islamic and Conventional Banks Stability in Indonesia

Khan,

Aziza,

Nawaz

2024

Etikonomi

View full text Add to dashboard Cite

show abstract

“…Bueff et al (2022) developed a data perturbation technique to generate stress scenarios using counterfactual metrics to assess the interpretability of machine learning credit scoring techniques. Nwafor and Nwafor (2023) used the shapley additive explanations (SHAP) framework to evaluate the model feature importance rankings. Florez-Lopez and Ramon-Jeronimo (2015) proposed an integrated approach based on merging decision trees to produce rule-based structures that can be directly used for managerial decision-making.…”

Section: Introductionmentioning

confidence: 99%

Credit risk prediction based on causal machine learning: Bayesian network learning, default inference, and interpretation

Liu,

Zhang,

Xiong

2024

Journal of Forecasting

View full text Add to dashboard Cite

The predictive and interpretable power of models is crucial for financial risk management. The purpose of this study was to perform credit risk prediction in a structured causal network with four stages—data processing, structural learning, parameter learning, and interpretation of inferences—and use six real credit datasets to conduct empirical research on the proposed model. Compared with traditional machine learning algorithms, we comprehensively explain the results of credit default through forward and reverse reasoning. We also compared our model with the post hoc interpretation models local interpretable model‐agnostic explanations (LIME) and shapley additive explanations (SHAP) to verify the interpretability of Bayesian networks. The experimental results show that the prediction performance of Bayesian networks is superior to traditional machine learning models and similar to the performance of ensemble models. Furthermore, Bayesian networks offer valuable insights into the interplay of features by considering their causal relationships and enable an assessment of how individual features influence the prediction outcome. In this study, what‐if analysis was performed to assess credit default probabilities under various conditions. This analysis provides decision‐makers with the necessary tools to make informed judgments about the risk profile of borrowers. Consequently, we consider Bayesian networks as a viable tool for credit risk prediction models in terms of prediction performance and interpretability.

show abstract