Corporate Bankruptcy Prediction Using Machine Learning
DOI:
https://doi.org/10.54097/heb14966Keywords:
Bankruptcy; Machines Learning; Feature Selection; Resampling; Voting Classifier.Abstract
As the advancement of machine learning, hundreds of methods have been proposed in solving corporate bankruptcy prediction problems. Facing with these options, decision-makers must decide the most effective methods. On the other hand, bankruptcy data usually involve high-dimensionality and extreme class imbalance, which may undermine the performance of classical models. This research designs and validates a systematic framework to address these issues, combining feature selection, sample rebalancing and machine learning model selection together. The study chooses five typical machine learning models, including Logistic Regression (LR), Support Vector machine (SVM), Decision Tree (DT), XGBoost (XGB) and Random Forest (RF). Also, this study designs the training and test data sets using a two-layered feature selection method, comparing two resampling methods and five classic machine learning models to form a final improved voting classifier with hyperparameters tuned. The experimental results confirm that XGB and RF, when combined with oversampling method, can provide the most robust prediction.
Downloads
References
[1] Daniela R, Mária B, Lucia J. Analysis of the construction industry in the Slovak Republic by bankruptcy model. Procedia - Social and Behavioral Sciences, 2016, 230: 298–306. https://doi.org/10.1016/j.sbspro.2016.09.038
[2] Altman E I. Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. The Journal of Finance, 1968, 23(4): 589–609.
[3] Barboza F, Kimura H, Altman E. Machine learning models and bankruptcy prediction. Expert Systems with Applications, 2017, 83: 405–417. https://doi.org/10.1016/j.eswa.2017.04.003
[4] Lu Y, Zeng N, Liu X, Yi S. A new hybrid algorithm for bankruptcy prediction using switching particle swarm optimization and Support Vector Machines. Discrete Dynamics in Nature and Society, 2015, 2015: Article 783262. https://doi.org/10.1155/2015/783262
[5] Härdle W, Lee Y-J, Schäfer D, Yeh Y-R. Variable selection and oversampling in the use of smooth Support Vector Machines for predicting the default risk of companies. Journal of Forecasting, 2009, 28(6): 512–534. https://doi.org/10.1002/for.1103
[6] Son H, Hyun C, Phan D, Hwang H J. Data analytic approach for bankruptcy prediction. Expert Systems with Applications, 2019, 138: 112816. https://doi.org/10.1016/j.eswa.2019.07.033
[7] Liang D, Tsai C F, Wu H T. The effect of feature selection on financial distress prediction. Knowledge-Based Systems, 2015, 73: 289–297. https://doi.org/10.1016/j.knosys.2014.10.010
[8] Chi D-J, Chu C-C. Artificial Intelligence in Corporate Sustainability: Using LSTM and GRU for Going Concern Prediction. Sustainability, 2021, 13(21): 11631. https://doi.org/10.3390/su132111631
[9] Fasano F, Adornetto C, Zahid I, La Rocca M, Montaleone L, Greco G, Cariola A. The dilemma of accuracy in bankruptcy prediction: a new approach using explainable AI techniques to predict corporate crises. European Journal of Innovation Management, 2024, 28(11): 1-22. https://doi.org/10.1108/EJIM-06-2024-0633
[10] D’Ercole A, Me G. A Novel Approach to Company Bankruptcy Prediction Using Convolutional Neural Networks and Generative Adversarial Networks. Machine Learning and Knowledge Extraction, 2025, 7(3): 63. https://doi.org/10.3390/make7030063
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
