Utilidad del Deep Learning en la predicción del fracaso empresarial en el ámbito europeo

Mariano Romero Martínez; Pedro Carmona Ibáñez; José Pozuelo Campillo

doi:10.46661/revmetodoscuanteconempresa.5172

Autores/as

Mariano Romero Martínez Universidad de Valencia https://orcid.org/0000-0002-4569-868X
Pedro Carmona Ibáñez Universidad de Valencia https://orcid.org/0000-0002-9979-2727
José Pozuelo Campillo Universidad de Valencia https://orcid.org/0000-0001-8768-2443

DOI:

https://doi.org/10.46661/revmetodoscuanteconempresa.5172

Palabras clave:

fracaso empresarial, Deep Learning, aprendizaje automático, ratios financieros, modelo de predicción

Resumen

En este trabajo pretendemos constatar la utilidad de las redes neuronales del Deep Learning en la predicción del fracaso empresarial, en particular, de las redes de alimentación hacia adelante (feedforward neuronal networks, en inglés). Se trata de una metodología caracterizada por proporcionar muy buenos resultados en términos de capacidad predictiva cuando se dispone de grandes tamaños muestrales. Para ello hemos desarrollado un modelo de predicción empresarial en empresas europeas, basado en dicho algoritmo, sobre una muestra formada por 61.624 empresas de las cuales 12.128 fueron declaradas en concurso en 2016. Como variables independientes se han considerado ratios y magnitudes económico-financieras obtenidas de las cuentas anuales del año anterior a la fecha del fracaso. Deep Learning logra una capacidad predictiva del 94%, de manera que presentan una mayor propensión al fracaso aquellas que tienen un mayor tamaño y una solvencia menor. Los resultados que se presentan se han contrastado en una submuestra de comprobación independiente y diferente a la empleada para estimar el modelo.

Descargas

Los datos de descargas todavía no están disponibles.

Citas

Abbott, A., Deshowitz, A., Murray, D., & Larson, E. (2018). WalkNet: A Deep Learning Approach to Improving Sidewalk Quality and Accessibility. SMU Data Science Review, 1(1-7), 1-12.

Ahn, B.S., Cho, S.S., & Kim, C.Y. (2000). The integrated methodology of rough set theory and artificial neural network for business failure prediction. Experts Systems with Aplications, 18, 65-74.

Alfaro, E., Gámez, M., & García, N. (2007a). A Boosting Approach for Corporate Failure Prediction. Applied Intelligence, 27(1), 29-37.

Alfaro, E., Gámez, M., & García, N. (2007b). Multiclass Corporate Failure Prediction by Adaboost.M1. International Advances in Economic Research, 13(3), 301-312.

Alfaro, E., Gámez, M., García, N., (2008). Linear Discriminant Analysis Versus Adaboost for Failure Forecasting. Revista Española de Financiación & Contabilidad, 37(137), 13-32.

Alfaro, E., García, N., Gámez, M., & Elizondo, D. (2008). Bankruptcy Forecasting: An Empirical Comparison of AdaBoost and Neural Networks. Decisión Support Systems, 45(1), 110-122.

Alfaro, E., Gámez, M., & García, N. (2013). Adabag: An R Package for Classification with Boosting and Bagging. Journal of Statistical Software, 54(2), 1-35.

Bell, T.B., Ribar, G.S., & Verchio, J. (1990). Neural Nets Versus Logistic Regression: A Comparison of Each Model’s Ability to Predict Commercial Bank Failures. En Srivastava, R.P. (ed). Proceedings of the 1990 Deloitte & Touche/University of Kansas Symposium in Auditing Problems, 29-53. Auditing Symposium X. Universidad de Kansas, Lawrence.

Boritz, J., & Kennedy, D.B. (1995). Effectiveness of Neuronal Network Types for prediction of Business Failure. Experts Systems with Aplications, 9(4), 503-512.

Brownlee, J. (2017). What is the Difference Between a Parameter and a Hyperparameter? https://machinelearningmastery.com/difference-between-a-parameter-and-a-hyperparameter/ (visitado el 28 de marzo de 2018)

Chaudhuri, A., & Ghosh, S. K. (2017). Bankruptcy Prediction through Soft Computing based Deep Learning Technique. Germany: Springer.

Cook, D. (2017). Practical Machine Learning with H2O. Sebastopol, CA (USA): O’Reilly Media.

Cortes, C., & Mohri, M. (2004). AUC optimization vs, error rate minimization, Advances in Neural Information Processing Systems, 16(16), 313-320.

Crawford, C. (2016). An Introduction to Deep Learning. https://blog.algorithmia.com/introduction-to-deep-learning-2016/ (visitado el 30 de marzo de 2018)

Deng, L., & Yu, D. (2013). Deep Learning: Methods and Applications. Foundations and Trends® in Signal Processing, 7(3-4), 197-387. http://dx.doi.org/10.1561/2000000039

Díaz, Z., Fernández, J., & Segovia, M.J. (2004). Sistemas de inducción de reglas y árboles de decisión aplicados a la predicción de insolvencias en empresas aseguradoras. Documentos de Trabajo de la Facultad de Ciencias Económicas y Empresariales, 9. Universidad Complutense de Madrid.

Fletcher, D., & Goss, E. (1993). Application Forecasting with Neural Networks: An Application Using Bankruptcy Data. Information and Management, 24, 159-167.

Friedman, J.H. (2002). Stochastic Gradient Boosting. Journal of Computational Statistics & Data Analysis, 38(4), 367-378.

Goodfellow, I., Bengio, Y., & Aaron, C., (2016). Deep Learning. Cambridge, MA, USA: The MIT Press.

Gulli, A., & Pal, S. (2017). Deep Learning with Keras. Implement neural networks with Keras on Theano and Tensorflow. Birmingham, Reino Unido: Packt Publishing Ltd.

Jayanthi, J., Kaur, G., & Suresh, K. (2017). Financial forecasting using decision tree (reptree & c4.5) and neural networks (k*) for handling the missing values. ICTAC Journal on Soft Computing, 7(3), 1473-1477.

Khashman, A. (2010). Neural networks for credit risk evaluation: Investigation of different neural models and learning schemes. Expert Systems with Applications, 37(9), 6233-6239.

Kim, M.J., & Kang, D.K. (2010). Ensemble with Neural Networks for Bankruptcy Prediction. Expert System with Applications, 37(4), 3373-3379.

Kim, M.J., Kang, D.K., & Kim, H.B. (2015). Geometric Mean Based Boosting Algorithm with Over-Sampling to Resolve Data Imbalance Problem for Bankruptcy Prediction. Expert Systems with Applications, 42(3), 1074-1082.

Kim, S.Y., & Upneja, A. (2014). Predicting Restaurant Financial Distrees Using Decisión Tree and AdaBoosted Decision Tree Models. Economic Modelling, 35, 354-362.

LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521, 436-444. http://dx.doi.org/10.1038/nature14539

Ledell, E. (2018). Deep Learning with H2O. https://github.com/ledell (visitado el 20 de marzo de 2018)

López, F.J., & Pastor, I. (2015). Bankruptcy visualization and prediction using neural netwoks: A study of U.S. comercial banks. Experts Systems with Aplications, 42, 2857-2869.

Momparler, A., Carmona, P., & Climent, F.J. (2016). La predicción del fracaso bancario con la metodología Boosting Classification Tree. Revista Española de Financiación y Contabilidad, 45(1), 63-91.

Natekin A., & Knoll, A. (2013). Gradient Boosting Machines. A Tutorial. Frontiers in Neurorobotics. 7(21), 1-21.

Odom, M.D., & Sharda, R. (1992). A Neural Network Model for Bankruptcy Prediction. En R.R. Trippi and E. Turban (Eds.). Neural networks in Finance and Investing. Chicago: Probus Publishing.

Patel, V., Armstrong, D., Ganguli, M.P., Roopa, S., Kantipudi, N., Aalbashir, S., & Kamarth, M. (2016). Deep Learning in Gastrointestinal Endoscopy. Critical Reviews™ in Biomedical Engineering, 44(6), 493-504. http://dx.doi.org/10.1615/CritRevBiomedEng.2017025035

Popescu, M.E., Andreica, M., & Popescu, I-P. (2017). Decision support solution to business failure prediction. Proceedings of the International Management Conference, Faculty of Management, Academy of Economic Studies, Bucharest, Romania, 11(1), 99-106.

Pozuelo, J., Martínez, J., & Carmona, P. (2018). Análisis de la utilidad del algoritmo Gradient Boosting Machine (GBM) en la predicción del fracaso empresarial. Spanish Journal of Finance and Accounting / Revista Española de Financiación y Contabilidad, 47(4) 507-532. http://dx.doi.org/10.1080/02102412.2018.1442039.

R Core Team (2019). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. Vienna, Austria. URL http://www.R-project.org/.

Ravi, P., & Ravi, V. (2007). Bankruptcy Prediction in Banks and Firms ViaStatistical and Intelligent Techniques - A Review. European Journal of Operational Research, 180(1), 1-28.

Romero, F. (2013). Alcances y limitaciones de los modelos de capacidad predictiva en el análisis del fracaso empresarial. AD-minister, 23, 45-70.

Schmidhuber, J. (2015). Deep learning in neural networks: An overview. Neural Networks, 61, 85-117.

Tam, K.Y. (1991). Neural Network Models and the Prediction of Bank Bankruptcy, Omega, 19(5), 429-445.

Tam, K.Y., & Kiang, M.Y. (1992). Managerial Applications of Neural Networks: The Case of Bank Failure Predictions. Management Science, 38(7), 926-947.

Tascón, M.T., & Castaño, F.J. (2012). Variables y modelos para la identificación y predicción del fracaso empresarial: revisión de la investigación empírica reciente. Revista de Contabilidad, 15(1), 7-58.

The H2O.ai team (2019). h2o: R Interface for H2O. R package version 3.26.0.2. https://CRAN.R-project.org/package=h2o.

Tseng, F-M., & Hu, Y-Ch. (2010). Comparing fou nakrupcty prediction models: Logit, quadratic interval logit, neural and fuzzy neural networks. Experts Systems with Aplications, 37, 1846-1853.

Wang, G., Ma, J., & Yang, S. (2014). An Improved Boosting Based on Feature Selection for Corporate Bankruptcy Prediction. Expert Systems with Applications, 41(5), 2353-2361.

Wang, L., Zeng, Y., & Chen, T. (2015). Back propagation neural network with adaptive differential evolution algorithm for time series forecasting. Expert Systems with Applications, 42, 855-863.

West, D., Dellana, S., & Qian, J. (2005). Neural Network Ensemble Strategies for Financial Decision Applications. Computers & Operations Research, 32(10), 2543-2559.

Wilson, G.I., & Sharda, R. (1994). Bankruptcy Prediction Using Neural Network. Decision Support Systems, 11, 545-557.

Zhang, G.P., Hu, M.Y., Patuwo, B.E., & Indro, D.C. (1999). Artificial Neural Networks in Bankruptcy Prediction: General Framework and Cross-Validation Analysis. European Journal of Operational Research, 116(1), 16-32.

Zieba, M., Tomczak, S.K., & Tomczak, J.M. (2016). Ensemble Boosted Trees with Synthetic Features Generation in Application to Bankruptcy Prediction. Expert Systems with Applications, 58(1), 93-101.