Liquidity Dominance and Nonlinear Stability Dynamics in Indonesian Islamic Banking Using Artificial Neural Networks
Journal of Business Management and Islamic Banking · 2026
DOI: 10.14421/jbmib.2026.0502-01 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Business Management and Islamic Banking, 2026. Verfasst von Iman Supriadi und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Research Aims: This study aims to examine the non-linear relationship between operational efficiency, profitability, and financing risk in Indonesian Islamic commercial banks, while developing predictive models for systemic risk and identifying key determinants of financial instability. Design/methodology/approach: The study adopts a quantitative explanatory approach, integrating descriptive statistics and Pearson correlation with Artificial Neural Networks (ANN) and Recurrent Neural Networks (RNN) to capture complex non-linear relationships. Model interpretability is enhanced using SHapley Additive exPlanations (SHAP) to assess the relative importance of predictors. Research Findings: The findings reveal that the relationship between operational efficiency and financing risk is inherently non-linear, characterized by a critical threshold beyond which efficiency improvements increase NPF. The ANN model demonstrates high predictive accuracy, confirming its effectiveness as an early warning system. Furthermore, liquidity conditions, represented by FDR, emerge as the dominant driver of systemic risk, supporting the Bad Luck hypothesis over the Bad Management hypothesis. Theoretical Contribution/Originality: This study introduces the concept of the Digital Efficiency Stability Threshold, which redefines efficiency as a bounded construct within Islamic banking. It also integrates machine learning and explainable AI into the analysis of financial stability, bridging the gap between predictive performance and interpretability. Research limitation and implication: The study is limited to a short observation period and a single-country context. Future research should incorporate longer time horizons, cross-country data, and additional macroeconomic variables to enhance generalizability and robustness.
Zitieren
APA
Supriadi, I., Maghfiroh, R. U., & Abadi, R. (2026). Liquidity Dominance and Nonlinear Stability Dynamics in Indonesian Islamic Banking Using Artificial Neural Networks. Journal of Business Management and Islamic Banking. https://doi.org/10.14421/jbmib.2026.0502-01
Vancouver
Supriadi I, Maghfiroh RU, Abadi R. Liquidity Dominance and Nonlinear Stability Dynamics in Indonesian Islamic Banking Using Artificial Neural Networks. Journal of Business Management and Islamic Banking. 2026. doi:10.14421/jbmib.2026.0502-01
BibTeX
@article{Supriadi2026,
author = {Supriadi, Iman and Maghfiroh, Rahma Ulfa and Abadi, Rukhul},
title = {Liquidity Dominance and Nonlinear Stability Dynamics in Indonesian Islamic Banking Using Artificial Neural Networks},
journal = {Journal of Business Management and Islamic Banking},
year = {2026},
doi = {10.14421/jbmib.2026.0502-01}
}
EUVIMED