The Application of Genetic Algorithms in Bankruptcy Predication and the Comparison of it with Altman's Z-model listed companies in Tehran Stocks Exchange (TSE)

Bankruptcy predication is one of the main matters in classifying the bankrupt corporations. Investors, Owners, Managers, Creditors, Governmental Agencies are interested in evaluating the financial conditions of the companies. Since in the case of bankruptcy, it will cost them a lot. This research intended to study the bankruptcy prediction in Tehran Stock Exchange (TSE) by using the Altman’s Z-model, Genetic Algorithms and finally intends to determine efficiency of the best model. The samples under the investigation were 36 bankrupt companies and 36 non-bankrupt companies during fiscal years of 1384 – 1386. The variables used in Genetic Algorithms Model and Altman’s Z- Model were 5 variables. The Genetic Algorithms Model had respectively an equivalent accuracy between 90, 91.5 percent in average during one year and two years before the indicator year (base year).The Altman‘s Z-Model had equivalent accuracy of 83.32 and 83.32 percent. According to the finding, Genetic Algorithms Model (GAM) was more accurate in predicting bankruptcy as a result; it is an appropriate means for predicting bankruptcy.