Algoritmos de Random Forest como alerta temprana para la predicción de insolvencias en empresas constructoras = Random Forest algorithms as early warning tools for the prediction of insolvencies in construction companies

José Ignacio Sordo Sierpe, Mercedes Del Rio Merino, Alvaro Pérez Raposo, Veronica Vitiello


doi:10.20868/ade.2021.4766

Full Text: PDF

Abstract


La preocupación de la Unión Europea por evitar que las empresas lleguen a un procedimiento de insolvencia motivó la promulgación de la Directiva (UE) 2019/1023 del Parlamento Europeo y del Consejo, y su transposición obligatoria a las regulaciones de los Estados miembros antes del 17 de julio de 2021. Esta Directiva establece que los deudores deben tener acceso a herramientas de alerta temprana para detectar situaciones de insolvencia inminente. Esta investigación tiene como objetivo contribuir al desarrollo de este tipo de herramientas de alerta temprana para un sector muy específico: la construcción residencial y no residencial. La metodología se ha dividido en dos fases, cada una con su propio objetivo específico: (1) seleccionar las variables predictoras que mejor puedan explicar el modelo (para ello se han utilizado técnicas estadísticas tradicionales); y (2) seleccionar los algoritmos que proporcionen la mayor precisión para el modelo de herramienta de alerta temprana entre cinco algoritmos Random Forest. El objetivo principal de esto es obtener señales de alerta con la suficiente antelación para poder detectar situaciones de insolvencia. El objetivo fundamental es lograr un modelo sin utilizar las cuentas de pérdidas y ganancias de las constructoras investigadas. Esto es así para evitar la falta de objetividad que pueden tener los ingresos y, por tanto, los resultados contables en este sector. Se obtuvieron porcentajes de precisión superiores al 85% tres años antes de que ocurriera la insolvencia utilizando únicamente ratios de balance. El principal valor es poder aplicar la herramienta de alerta temprana de forma sencilla, utilizando pequeñas cantidades de datos, especialmente para el deudor, que puede reaccionar con la suficiente antelación para evitar una situación financiera potencialmente irreversible.

Abstract

The European Union's concern with preventing companies from reaching insolvency proceedings motivated the enactment of Directive (EU) 2019/1023 of the European Parliament and of the Council, and its mandatory transposition into Member States' regulations by July 17, 2021. This Directive states that debtors must have access to early warning tools to detect situations of imminent insolvency. This research aims to contribute to the development of such early warning tools for a very specific sector: residential and non-residential construction. The methodology has been divided into two phases, each with its own specific objective: (1) to select the predictor variables that can best explain the model (traditional statistical techniques have been used for this purpose); and (2) to select the algorithms that provide the greatest precision for the early warning tool model from among five Random Forest algorithms. The main objective of this is to obtain warning signs sufficiently enough in advance that insolvency situations can be detected. The fundamental aim is to achieve a model without using the profit and loss accounts from the construction companies under investigation. This is so to avoid the lack of objectivity that income, and therefore accounting results, may have in this sector. Accuracy percentages of over 85% were obtained three years before insolvency occurred using only balance sheet ratios. The main value is to be able to apply the early warning tool in a simple way, using little amounts of data, especially for the debtor, who can react early enough to avoid a potentially irreversible financial situation.


Keywords


Advertencia temprana; Random Forest; construcción; Early warning; Random Forest; construction.

References


Alaka, H. A. (2018). Systematic review of bankruptcy prediction models: Towards a framework for tool selection. Expert System with Applications, 164-184.

Altman, E. (1968). Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. The journal of finance, 589-610.

Altman, E. (2005). An emerging market credit scoring system for corporate bonds. Emerging markets review, 311-323.

Altman, E., Iwanicz-Drozdowska, M., Laitinen, E., & Suvas, A.(2017). Financial distress prediction in an internationalcontext. A review and empirical analysis of Altman's Z-score model. Journal of International Financial Management & Accounting, 131-171.

Arora, I. (2020). Prediction of corporate bankruptcy using financial ratios and news. International Journal of Engineering and Management Research, 82-87.

Beaver, W. (1966). Financial ratios as predictors of failure: An empirical research in accounting selected studies. Journal of Accounting Research, 71-111.

Becerra-Vicario, R., Alaminos, D., Aranda, E., & Fernández-Gámez, M. A. (2020). Deep recurrent convolutional neuronal network for bankruptcy prediction: A case of the restaurant industry. Sustainability, 1-15.

Beneish, M. (1999). The D¡detection of earningsmanipulation. Financial Analysts Journal, 24-36.

Blum, M. (1974). Failing company discriminant analysis. Journal of accounting research, 1-25.

Deakin, E. (1972). A discriminant analysis of predictor of business failure. Journal of accounting research, 167-179.

Edmister, R. (1972). An empirical test of financial ratio analysis por small business failure prediction. Journal of Financial and Quantitative Analysis, 1477-1493.

Garcia Catellví, A. (2005). El cálculo del resultado de los proyectos desarrollados por constructoras. Partida Doble, 42-53.

Tukey, J. W. (1977). Exploratory data analysis. Massachusset: Audison-Wesley Publishing Company.

Lantz, B. (2013). Machine Learning with R. Birmingham: Packt Publishing, Ltd.

Meyer, P. A., & Pifer, H. W.(1970). Prediction of bank failures. The Journal of Finance, 853-868.

Ohlson, J. (1980). Financial ratios and the `probabilistic of bankruptcy. Journal of accounting research, 109-131.

Perboli, G., & Arabnezhad, E. (2021). A Machine Learning-based DSS for mid and long-term company crisis prediction. Expert Systems with Applications, 1-12.

Rosner, R. (2003). Earnings manipulations in failing firms. Contemporary accounting research, 361-408.

Sajjan, R. (2016). Predicting bankruptcy of selected firms by applying Altman's Z-score model. International Journal of Research-Granthaalayah, 152-158.

Tabbakh, A., Kumar, J. S., & Janjhi, N. (2021). Bankruptcy Prediction using Robust Machine Learning Model. Turkish Journal of Computer and Mathematics Education, 3060-3073.

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

Wang, H., & Liu, X. (2021). Undersampling bankruptcy prediction: Taiwan bankruptcy data. Plos One, 1-17.

Zmijewski, M. (1984). Methodological Issues related to the estimation of financial distress prediction models. Journal of Accounting Research, 59-82.




Copyright (c) 2022 Autor / BY-NC

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.