This article addresses the financial aspects surrounding the stability of the electricity sector. We apply a variety of data mining techniques to build financial distress warning models based on the financial statement analysis method. The analysis reveal that neural networks with the accuracy of 80% and above in different scenarios were found to be relatively more accurate compared to decision trees and support vector machines. Additionally, in order to assess the ability of financial indicators, we applied feature selection. The financial ratios analyses proved the significance of profitability, liquidity and financial leverage for the default prediction models. Therefore, it is exigent that companies utilize their assets, liquidity and solvency as the core of their management policy regulations. The key contribution of this paper is the formulation of a proper model for financial distress prediction among electricity sector companies in Iran.
CITATION STYLE
Mirzaei, M., Hosseini, S. M. P., Gan, G. G., & Sahu, P. K. (2018). Prediction of financial distress for electricity sectors using data mining. In Lecture Notes on Data Engineering and Communications Technologies (Vol. 5, pp. 3–11). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-319-59427-9_1
Mendeley helps you to discover research relevant for your work.