Predicting Corporate Financial Distress Using Financial Ratios and Explainable Machine Learning: Evidence from Listed Companies
DOI:
https://doi.org/10.69980/bma.v12i3.2533Keywords:
Corporate financial distress, Explainable machine learning, Financial ratios, XGBoost, SHAPAbstract
The ability to predict corporate financial distress is crucial for investors, creditors, managers, and regulators, as it aids in timely intervention and risk management in firms with financial distress. This study tackles the problem of obtaining high predictive performance and model transparency. A machine-learning framework was built that is explainable, with Logistic Regression, Random Forest, XGBoost and LightGBM models, and stratified development, validation, and locked-test partitions. Analysis was performed on the UCI Taiwanese Bankruptcy Prediction dataset, which consists of 6,819 observations and 95 financial indicators of listed companies. The assessment of model performance was done by ROC-AUC, PR-AUC, precision, recall, specificity, F1-score, balanced accuracy, Matthews correlation coefficient, and Brier score; and SHAP was used for global and nonlinear interpretation. XGBoost performed the best overall with ROC-AUC = 0.958, PR-AUC = 0.602, MCC = 0.612, and Brier score = 0.022. Net Income to Total Assets, Debt Ratio, Persistent EPS, and ROA were the most significant predictors. The study provides an understandable financial-distress framework that is accurate, explainable, and risk-stratifies companies for decision support. Most distress cases were observed in a small, high-risk subgroup, along with risk stratification, which further prioritized early-warning cases in a practical way and allocated resources.
Downloads
References
J. Zhao, J. Ouenniche, and J. De Smedt, “Survey, classification and critical analysis of the literature on corporate bankruptcy and financial distress prediction,” Machine Learning with Applications, vol. 15, Art. no. 100527, 2024, doi: 10.1016/j.mlwa.2024.100527.
A. Dasilas and A. Rigani, “Machine learning techniques in bankruptcy prediction: A systematic literature review,” Expert Systems with Applications, vol. 255, Art. no. 124761, 2024, doi: 10.1016/j.eswa.2024.124761.
D. Kuizinienė, T. Krilavičius, R. Damaševičius, and R. Maskeliūnas, “Systematic review of financial distress identification using artificial intelligence methods,” Applied Artificial Intelligence, vol. 36, no. 1, Art. no. 2138124, 2022, doi: 10.1080/08839514.2022.2138124.
N. Nazareth and Y. V. R. Reddy, “Financial applications of machine learning: A literature review,” Expert Systems with Applications, vol. 219, Art. no. 119640, 2023, doi: 10.1016/j.eswa.2023.119640.
H. A. Alaka, L. O. Oyedele, H. A. Owolabi, V. Kumar, S. O. Ajayi, O. O. Akinade, and M. Bilal, “Systematic review of bankruptcy prediction models: Towards a framework for tool selection,” Expert Systems with Applications, vol. 94, pp. 164–184, 2018, doi: 10.1016/j.eswa.2017.10.040.
F. Barboza, H. Kimura, and E. Altman, “Machine learning models and bankruptcy prediction,” Expert Systems with Applications, vol. 83, pp. 405–417, 2017, doi: 10.1016/j.eswa.2017.04.006.
E. I. Altman, M. Iwanicz-Drozdowska, E. K. Laitinen, and A. Suvas, “A race for long horizon bankruptcy prediction,” Applied Economics, vol. 52, no. 37, pp. 4092–4111, 2020, doi: 10.1080/00036846.2020.1730762.
N. Almaskati, R. Bird, D. Yeung, and Y. Lu, “A horse race of models and estimation methods for predicting bankruptcy,” Advances in Accounting, vol. 52, Art. no. 100513, 2021, doi: 10.1016/j.adiac.2021.100513.
Q. Zhao, W. Xu, and Y. Ji, “Predicting financial distress of Chinese listed companies using machine learning: To what extent does textual disclosure matter?” International Review of Financial Analysis, vol. 89, Art. no. 102770, 2023, doi: 10.1016/j.irfa.2023.102770.
M. E. Lokanan and S. Ramzan, “Predicting financial distress in TSX-listed firms using machine learning algorithms,” Frontiers in Artificial Intelligence, vol. 7, Art. no. 1466321, 2024, doi: 10.3389/frai.2024.1466321.
D. Wu, X. Ma, and D. L. Olson, “Financial distress prediction using integrated Z-score and multilayer perceptron neural networks,” Decision Support Systems, vol. 159, Art. no. 113814, 2022, doi: 10.1016/j.dss.2022.113814.
Z. Zhang, C. Wu, S. Qu, and X. Chen, “An explainable artificial intelligence approach for financial distress prediction,” Information Processing & Management, vol. 59, no. 4, Art. no. 102988, 2022, doi: 10.1016/j.ipm.2022.102988.
P. Carmona, A. Dwekat, and Z. Mardawi, “No more black boxes! Explaining the predictions of a machine learning XGBoost classifier algorithm in business failure,” Research in International Business and Finance, vol. 61, Art. no. 101649, 2022, doi: 10.1016/j.ribaf.2022.101649.
H. Qian, B. Wang, M. Yuan, S. Gao, and Y. Song, “Financial distress prediction using a corrected feature selection measure and gradient boosted decision tree,” Expert Systems with Applications, vol. 190, Art. no. 116202, 2022, doi: 10.1016/j.eswa.2021.116202.
S. Deng, Q. Luo, Y. Zhu, H. Ning, and T. Shimada, “Financial risk forewarning with an interpretable ensemble learning approach: An empirical analysis based on Chinese listed companies,” Pacific-Basin Finance Journal, vol. 85, Art. no. 102393, 2024, doi: 10.1016/j.pacfin.2024.102393.
H. H. Nguyen, J.-L. Viviani, and S. Ben Jabeur, “Bankruptcy prediction using machine learning and Shapley additive explanations,” Review of Quantitative Finance and Accounting, vol. 65, no. 1, pp. 107–148, 2025, doi: 10.1007/s11156-023-01192-x.
S. Ben Jabeur, N. Stef, and P. Carmona, “Bankruptcy prediction using the XGBoost algorithm and variable importance feature engineering,” Computational Economics, vol. 61, no. 2, pp. 715–741, 2023, doi: 10.1007/s10614-021-10227-1.
[18] Q. Meng, X. Zheng, and S. Wang, “Corporate governance and financial distress in China: A multi-dimensional nonlinear study based on machine learning,” Pacific-Basin Finance Journal, vol. 88, Art. no. 102549, 2024, doi: 10.1016/j.pacfin.2024.102549.
F.-J. Zhu, L.-J. Zhou, M. Zhou, and F. Pei, “Financial distress prediction: A novel data segmentation research on Chinese listed companies,” Technological and Economic Development of Economy, vol. 27, no. 6, pp. 1413–1446, 2021, doi: 10.3846/tede.2021.15337.
T.-K. Chen, H.-H. Liao, G.-D. Chen, W.-H. Kang, and Y.-C. Lin, “Bankruptcy prediction using machine learning models with the text-based communicative value of annual reports,” Expert Systems with Applications, vol. 233, Art. no. 120714, 2023, doi: 10.1016/j.eswa.2023.120714.
S. Ding, T. Cui, A. G. Bellotti, M. Z. Abedin, and B. Lucey, “The role of feature importance in predicting corporate financial distress in pre- and post-COVID periods: Evidence from China,” International Review of Financial Analysis, vol. 90, Art. no. 102851, 2023, doi: 10.1016/j.irfa.2023.102851.
S. Ben Jabeur and V. Serret, “Bankruptcy prediction using fuzzy convolutional neural networks,” Research in International Business and Finance, vol. 64, Art. no. 101844, 2023, doi: 10.1016/j.ribaf.2022.101844.
UCI Machine Learning Repository, “Taiwanese Bankruptcy Prediction,” 2020, doi: 10.24432/C5004D.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
