Support Vector Machines Application in Financial Distress Prediction of Companies Using Financial Ratios

The development of the bankruptcy or financial distress prediction model has long been regarded as an important research in the academic and business entities. Financial distress of companies imposes many costs to the companies. One method that can help companies to prevent from financial distress is prediction of financial distress. This prediction also can help banks and other financial institution to have better credit scoring and rating systems. In this study we used Support Vector Machines (SVM) for predicting financial distress of companies and Logistic Regression (LR) as a comparative method. We found that SVM has a better performance than LR. Results show that SVM not only has a better accuracy rate of prediction but also has a better generalization power.