A Model to Predict Bankruptcy using the Mechanisms of Corporate Governance and financial ratios

Improving the economic and business environment is the most important factor in preventing bankruptcy, therefor, Artificial intelligence uses to predict the bankruptcy of companies in the future. In this study, companies in the Tehran Stock Exchange over a period of 10 years in terms of bankruptcy localized model of Kurdistani-Tatli based on the Altman model were examined and companies were identified as bankrupt and healthy. Research data were collected, categorized and refined using secondary data extracted from financial statements and through the database of the Exchange Organization and the Central Bank.The models used to evaluate the data and predict the bankruptcy of companies are artificial intelligence models . Artificial neural network, combination of neural network and genetic algorithm and the K-nearest neighbor method has been used. They were also compared in terms of prediction accuracy. The output of the models indicates that the addition of corporate governance indicators to the financial ratios indicators has not improved the results. Therefore, financial ratios alone are sufficient for predicting and determining bankruptcy. The proposed model of this research based on accuracy is a combined model of neural network and genetic algorithm that has the highest accuracy. Genetic algorithm improves the optimal results of the neural network and provides a more optimal answer.