Predicting corporate bankruptcy using Artificial Neural Networks (ANN) in Tehran Stock Exchange (TSE

The main purpose of this paper is prediction of TSE corporate financial bankruptcy using Artificial Neural Networks. The mean values of key ratios reported in past bankruptcy studies were selected for neural network inputs (Working capital to total assets, Net income to total assets, Total debt to total assets, Current assets to current liabilities, Quick assets to current liabilities). The neural network used in this paper is Multilayer Perceptron (MLP) that trained with back propagation algorithm, and contained three-layer feed forward neural network with 5,4,1 number of neurons in input, hidden and output layer respectively. The samples of this paper consist of bankrupt and non-bankrupt groups. Bankruptcy group was Manufacturing Corporations that were included Article 141 of Mercantile law within 1378-1385 and non-bankruptcy group selected by random sampling. The same set of data is analyzed using more traditional method of bankruptcy prediction, multivariate discriminate analysis. A comparison of the predictive abilities of both the neural network and the discriminate analysis method was presented. Also, prediction accuracy of neural network is presented by ROC curve. The results showed that there is significant difference between MDA and ANN, Also, according to the results, lower error type 1 has priority to error type 2.