Bankruptcy Prediction of Indian Companies Using Neural Network

DOI: 10.5176/2251-1997_AF16.30

Authors: Abhay Nagale and Varadraj Bapar

Abstract:

Bankruptcy is a serious threat to any organization and so there is a need for empirical models which can predict bankruptcy. In recent years machine learning techniques have been used to develop bankruptcy prediction model. In this study back-propagation neural network was applied to a sample of bankrupt and non-bankrupt Indian companies. Models were developed for years one, two and three prior to bankruptcy. The predictive accuracy of the model was 91 per cent one year prior to bankruptcy. This model could also predict 80 per cent of the cases three years prior to bankruptcy.

Keywords: bankruptcy prediction, artificial intelligence, neural network