Abstract
The purpose of the paper is to propose two new procedures that deal with overfitting problem using neural techniques for variable selection and business failure prediction. The first procedure, called HVS-AUC, is based simultaneously on (i) the backward search, (ii) the HVS measure (Heuristic for Variable Selection), and (iii) the AUC criterion (Area Under Curve). The second procedure, called forward-AUC, is based on (i) the forward search and (ii) the AUC criterion. Using a sample of bankrupt and non-bankrupt firms in France, the implementation of the procedures using neural networks shows that the profitability, the repayment capacity, the taxation, and the importance of investment have a strong explanatory power in bankruptcy prediction. These procedures also provide more parsimonious and more efficient models compared to Linear Discriminant Analysis.
Similar content being viewed by others
Notes
Overfitting problem occurs when the model fits the training data very well but has poor predictive performance on a new dataset.
The HVS measure assess the pertinence of input variable to the output decision (Yacoub and Bennani 2000). It will be presented in step 2.
Details explaining which ratios were grouped into which category and the definition of each ratio are reported in “Appendix A”.
References
Akers, M., Bellovary, J., & Giacomino, D. (2007). A review of bankruptcy prediction studies: 1930 to present. Journal of Financial Education, 33(5), 1–42.
Alfaro, E., Garcia, N., Games, M., & Ellison, D. (2008). Bankruptcy forecasting: An empirical comparison of AdaBoost and neural networks. Decision Support Systems, 45, 110–122.
Altman, E. I. (1968). Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. Journal of Finance, 23(4), 589–609.
Altman, E., Halderman, R., & Narayanan, P. (1977). Zeta analysis. A new model to identify bankruptcy risk of corporations. Journal of Banking and Finance, 10, 29–54.
Back, B., Oosterom, G., Sere, K., & van Wezel, M. (1996a). Choosing bankruptcy predictors using discriminant analysis, logit analysis and genetical gorithms. Technical Report. Turku Centre for Computer Science
Back, B., Oosterom, G., & Sere, K. (1996b). Neural networks and genetical gorithms for bankruptcy predictions. Expert Systems with Applications, 11(4), 407–413.
Bapat, V., & Nagale, A. (2014). Comparison of bankruptcy prediction models: Evidence from India. Accounting and Finance Research, 4(3), 91–98.
Bateni, L., & Asghari, F. (2020). Bankruptcy prediction using logit and genetic algorithm models: A comparative analysis. Computational Economics, 55(1), 335–348.
Beaver, W. H. (1966). Financial ratios as predictors of failure. Journal of Accounting Research, 4, 71–102.
Bose, I., & Pal, R. (2006). Predicting the survival or failure of click-and-mortar corporations: A knowledge discovery approach. European Journal of Operational Research, 174, 959–982.
Brabazon, A., & Keenan, P. B. (2004). A Hybrid genetic model for the prediction of corporate failure. Computational Management Science, Special Isuue, 1(3–4), 293–310.
Canbas, S., Cabuk, A., & Kilic, S. B. (2005). Prediction of commercial bank failure via multivariate statistical analysis of financial structures: The Turkish case. European Journal of Operational Research, 166, 528–546.
Charalambous, C., Charitou, A., & Kaourou, F. (2000). Comparative analysis of artificial neural network models: Application in bankruptcy prediction. Annals of Operations Research, 99, 403–425.
Charitou, A., Neophytou, E., & Charalambous, C. (2004). Predictingcorporatefailure: Empirical evidence for the UK. European Accounting Review, 13, 465–497.
Doumpos, M., & Zopounidis, C. (2007). Model combination for credit risk assessment: A stacked generalization approach. Annals of Operations Research, 151, 289–306.
Du Jardin, P. (2010). Predicting bankruptcy using neural networks and other classification methods: The influence of variable selection techniques on model accuracy. Neurocomputing, 73, 2047–2060.
Du Jardin, P. (2012). The influence of variable selection methods on the accuracy of bankruptcy prediction models. Bankers, Markets & Investors, 116, 20–39.
Giordano, F., La Rocca, M., & Perna, C. (2014). Input variable selection in neural network models. Communications in statistics–Theory and Methods, 43(4), 735–750.
Huang, Z., Chen, H., Hsu, C. J., Chen, W. H., & Wu, S. (2004). Credit rating analysis with support vector machines and neural networks: A market comparative study. Decision Support Systems, 37, 543–558.
Jo, H., Han, I., & Lee, H. (1997). Bankruptcy prediction using case-based reasoning, neural networks and discriminant analysis. Expert Systems with Applications, 13(2), 97–108.
Kim, M. J., & Han, I. (2003). The discovery of experts’ decision rules from qualitative bankruptcy data using genetic algorithms. Expert Systems with Applications, 25(4), 637–646.
Kiviluoto, K. (1998). Predicting bankruptcy with the Self-Organizing Map. Neurocomputing, 21(1–3), 191–220.
Lee, K. C., Han, I., & Kwon, Y. (1996). Hybrid neural network models for bankruptcy predictions. Decision Support Systems, 18(1), 63–72.
Leray, P., & Gallinari, P. (2001). De l’utilisation d’OBD pour la selection de variables dans les perceptrons multicouches. Revue d’Intelligence Artificielle, 15, 373–391.
Leray, P., & Gallinari, P. (1998). Feature selection with neural networks. Behaviormetrika, 26, 145–166.
Li, M.-Y.L., & Miu, P. (2010). A hybrid bankruptcy prediction model with dynamic loadings on accounting-ratio-based and market-based information: A binary quantile regression approach. Journal of Empirical Finance, 17(4), 818–833.
Min, J. H., & Lee, Y. C. (2005). Bankruptcy prediction using support vector machine with optimal choice of kernel function parameters. Expert Systems with Applications, 28, 603–614.
Min, S. H., Lee, J., & Han, I. (2006). Hybrid genetic algorithms and support vector machines for bankruptcy prediction. Expert Systems with Applications, 31(3), 652–660.
Naidu, G., & Govinda, K. (2018). Bankruptcy prediction using neural networks. In Proceedings of the 2nd international conference on inventive systems and control, ICISC 2018 (pp. 248–251). https://doi.org/10.1109/ICISC.2018.8399072.
Ohlson, J. A. (1980). Financial ratios and the probabilistic prediction of bankruptcy. Journal of Accounting Research, 18(1), 109–131.
Poddig, T. (1995). Bankruptcy prediction: A comparison with discriminant analysis. In A. Refenes (Ed.), Article in neural network in the capital markets (pp. 311–323). Chichester: Wiley.
Sexton, R. S., Sriram, R. S., & Etheridge, H. (2003). Improving decision effectiveness of artificial neural networks: A modified genetic algorithm approach. Decision Sciences, 34(3), 421–442.
Tam, K. (1991). Neural network models and the prediction of bankruptcy. Omega, 19(5), 429–445.
Tung, W. L., Quek, C., & Cheng, P. (2004). A novel neural-fuzzy based early warning system for predicting bank failures. Neural Networks, 17(4), 567–587.
Veganzones, D., & Severin, E. (2020). Corporate failure prediction models in the twenty-first century: A review. European Business Review. https://doi.org/10.1108/EBR-12-2018-0209
Wilks, S. S. (1963). Mathematical statistics. New York: Wiley.
Wilson, R., & Sharda, R. (1994). Bankruptcy prediction using neural networks. Decision Support Systems, 11(5), 545–557.
Yacoub, M., & Bennani, Y. (2000). Features selection and architecture optimization in connectionist systems. International Journal of Neural Systems, 10(5), 379–395.
Yang, Z., & Harrison, R. G. (2002). Analysing company performance using templates. Intelligent Data Analysis, 6, 3–15.
Zhang, G., Hu, M., Patuwo, B., & Indro, D. (1999). Artificial neural networks in bankruptcy prediction: General framework and cross-validation analysis. European Journal of Operational Research, 116(1), 16–32.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
1.1 Appendix A: The financial ratios used in the analysis
Firm’s activity ratios
R1 Export rate (Export/Turnover)
R2 Stored production rate (Stored production (N) - Stored production (N-1)/Stored production (N-1))
R3 Value added ratio
R4 Rate of sold services in the turnover
R5 Rate of sold goods in the turnover
R6 Deadline cycle of goods in stock and under production
Operating ressources ratios
R7 Cost of labor
R8 Investment/Global added value
R9 Share of intangible assets relative to the total amount of fixed assets
R10 Share of other assets relative to the total amount of fixed assets
Operating cycle ratios
R11 Working capital needs in days of turnover
R12 Share of other accounts and debt receivable in the working capital needs in days of turnover
R13 Day’s receipt in advance ratio
R14 Trade debt/purchase and outsourcing costs
Earnings and margins ratios
R15 Profitability (Net income/Total assets)
R16 Return on equity
Financial structure ratios
Solvency ratios
R17 Financial costs/global production
R18 Leverage ratio
R19 Uncertain debt/sales
Leverage ratios
R20 Trade debt/total assets
R21 Lease liabilities/financial debt
R22 Current assets/invested capital
R23 Internal financing
Other ratios related to the financial structure
R24 Income tax and taxes other than income tax/Global added value
R25 Interest/EBITDA
R26 Interest/turnover
R27 Short-term banking debt/financial debt
R28 Banking debt/financial debt
Growth ratios
R29 Production growth rate
R30 Invested capital growth rate
Rights and permissions
About this article
Cite this article
Abid, I., Ayadi, R., Guesmi, K. et al. A new approach to deal with variable selection in neural networks: an application to bankruptcy prediction. Ann Oper Res 313, 605–623 (2022). https://doi.org/10.1007/s10479-021-04236-4
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10479-021-04236-4