Studying on Forecasting the Enterprise Bankruptcy Based on SVM

Enterprise bankruptcy forecasting is very important to manage credit risk, and a lot of scholars applied themselves to study how to increase the accuracy of bankruptcy forecast. The paper introduces a new bankruptcy forecast tool: support machine vector (SVM). SVM is based on the Vapnik-Chervonenks (VC) dimension of statistical learning theory and structural risk minimization inductive principle, which can deal with the small sample, non-linear and high dimension problem, and some researchers applied the SVM model to the artificial intelligence identification and biology identification. The paper introduces the theory of SVM, the regression algorithm and implement of SVM, then use the data of China Stock Exchange to analyze positively, the results show that the SVM model has the higher accuracy compared with neural net or common statistical methods when forecasting enterprise bankruptcy