Using Genetic Algorithm and Support Vector Machines for bankruptcy prediction: Empirical observation in Iran

The high social costs associated with bankruptcy have spurred for better theoretical understanding and prediction capability. In this paper, we investigated application of Genetic Algorithm (GA) and Support Vector Machines (SVM) in predicting companies' bankruptcy. So doing, having read the review of literature, the researchers found a complete list of financial proportions that showed high capabilities the predicting bankruptcy. These proportions include the ratio of operational income to sale, ratio of total debts of total assets; current assets to current debts; sale to current assets and interest cost to grass profit. Then, GA and SVM were applied to classify 132 bankrupt and non-bankrupt Iranian firms listed in Tehran stock exchange (TSE) for the period 2004 to 2007. McNemar test showed that GA approach outperforms SVM to the problem of predicting companies' bankruptcy.