The prediction of Bankruptcy Risk Investigation Using Artificial Neural Networks Based on Multilayer Perceptron Approach (Empirical Evidence: Tehran Stock Exchange)
Subject Areas : FuturologySomayeh Saroei 1 , Hamid Reza Vkili Fard 2 , Ghodratolah Taleb Nia 3
1 - Department of Accounting, Faculty of Basic Sciences, Kangavar Islamic Azad University, Kangavar, Iran.
2 - Department of Accounting, Faculty of Economics, Islamic Azad University, Science and Research Branch of Tehran, Tehran, Iran.
3 - Department of Accounting, Faculty of Economics, Islamic Azad University, Science and Research Branch of Tehran, Tehran, Iran.
Keywords: Artificial Neural Networks, Bankruptcy Risk, Bankruptcy Prediction,
Abstract :
The aim of this research is Identification of the effective factors on bankruptcy prediction of Iranian companies by findings of artificial neural network (ANN) system based on Multilayer Perceptron Approach (PS) , and providing an appropriate statistical model for estimating the bankruptcy of Iranian companies by using the findings of The ANN implementation. we seek to answer the following question: Are we able to design a valid statistical model by using findings of artificial neural network (ANN) system to predict the bankruptcy of Iranian companies? The statistical population in this study is all of listed companies in Tehran Stock Exchange. By considering the criteria and method of systematic deletion, 172 companies from this statistical society have been selected as the sample in this research from 2007 to 2016. In order to make statistical analyzes in this study, we used from methods such as artificial neural network system based on multilevel perceptron approach, binary logistic regression, and tests such as Akaic, Schwarz, Hanan Quinn and Z wang test. The results of the analysis of the research data show that the ANN system can identify of the factors affecting on bankruptcy of Iranian companies in the year before bankruptcy by Precision equal 98%.
_||_
