Skip to main content
Log in

A new approach to deal with variable selection in neural networks: an application to bankruptcy prediction

  • S.I.: Risk Management Decisions and Value under Uncertainty
  • Published:
Annals of Operations Research Aims and scope Submit manuscript

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.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Diagram 1
Diagram 2
Diagram 3
Diagram 4
Diagram 5
Diagram 6
Diagram 7
Diagram 8
Diagram 9
Diagram 10

Similar content being viewed by others

Notes

  1. Overfitting problem occurs when the model fits the training data very well but has poor predictive performance on a new dataset.

  2. The HVS measure assess the pertinence of input variable to the output decision (Yacoub and Bennani 2000). It will be presented in step 2.

  3. 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.

    Google Scholar 

  • 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.

    Article  Google Scholar 

  • Altman, E. I. (1968). Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. Journal of Finance, 23(4), 589–609.

    Article  Google Scholar 

  • 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.

    Article  Google Scholar 

  • 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.

    Article  Google Scholar 

  • Bapat, V., & Nagale, A. (2014). Comparison of bankruptcy prediction models: Evidence from India. Accounting and Finance Research, 4(3), 91–98.

    Google Scholar 

  • Bateni, L., & Asghari, F. (2020). Bankruptcy prediction using logit and genetic algorithm models: A comparative analysis. Computational Economics, 55(1), 335–348.

    Article  Google Scholar 

  • Beaver, W. H. (1966). Financial ratios as predictors of failure. Journal of Accounting Research, 4, 71–102.

    Article  Google Scholar 

  • 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.

    Article  Google Scholar 

  • 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.

    Google Scholar 

  • 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.

    Article  Google Scholar 

  • Charitou, A., Neophytou, E., & Charalambous, C. (2004). Predictingcorporatefailure: Empirical evidence for the UK. European Accounting Review, 13, 465–497.

    Article  Google Scholar 

  • Doumpos, M., & Zopounidis, C. (2007). Model combination for credit risk assessment: A stacked generalization approach. Annals of Operations Research, 151, 289–306.

    Article  Google Scholar 

  • 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.

    Article  Google Scholar 

  • Du Jardin, P. (2012). The influence of variable selection methods on the accuracy of bankruptcy prediction models. Bankers, Markets & Investors, 116, 20–39.

    Google Scholar 

  • 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.

    Article  Google Scholar 

  • 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.

    Article  Google Scholar 

  • 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.

    Article  Google Scholar 

  • Lee, K. C., Han, I., & Kwon, Y. (1996). Hybrid neural network models for bankruptcy predictions. Decision Support Systems, 18(1), 63–72.

    Article  Google Scholar 

  • 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.

    Article  Google Scholar 

  • 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.

    Article  Google Scholar 

  • 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.

    Article  Google Scholar 

  • 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.

    Article  Google Scholar 

  • 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.

    Google Scholar 

  • 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.

    Article  Google Scholar 

  • Tam, K. (1991). Neural network models and the prediction of bankruptcy. Omega, 19(5), 429–445.

    Article  Google Scholar 

  • 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.

    Article  Google Scholar 

  • 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

    Article  Google Scholar 

  • Wilks, S. S. (1963). Mathematical statistics. New York: Wiley.

    Google Scholar 

  • Wilson, R., & Sharda, R. (1994). Bankruptcy prediction using neural networks. Decision Support Systems, 11(5), 545–557.

    Article  Google Scholar 

  • Yacoub, M., & Bennani, Y. (2000). Features selection and architecture optimization in connectionist systems. International Journal of Neural Systems, 10(5), 379–395.

    Article  Google Scholar 

  • Yang, Z., & Harrison, R. G. (2002). Analysing company performance using templates. Intelligent Data Analysis, 6, 3–15.

    Article  Google Scholar 

  • 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.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Ilyes Abid.

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

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

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

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10479-021-04236-4

Keywords

Navigation