Application Of Support Vector Machine In Listing Corporation’s Financial Crisis

Many scholars concern about financial distress prediction of enterprise since the middle of the twentieth century. With the rapid development of China’s economy and the reform of the economic system, enterprises are responsible for their own profits and losses. They face the increasing opportunities and risks. After the system of entry and exit to the security market became more reasonable, enterprises with continuous loss will be treated as “ST” or “*ST”, till be delisted. Therefore, the establishment of financial distress can not only solve the existence crisis, but also avoid causing enormous losses to investors and creditors.Based on the findings concerning listed enterprises financial distress studied by Chinese and foreign scholars, we choose 150 listed enterprises’ financial datum which includes 16 financial indexes and 4 non-financial indexes. We combine the theory of financial distress warning with support vector machine to build the model. In order to forecast the special treatment of companies better, we also optimize the model using the particle swarm optimization and grid search algorithm. And then we compare prediction results with the test sample datum, we will find the SVM has great accuracy.In order to better reflect the advantage of SVM in the prediction accuracy of financial distress situation, we use optimized SVM、BP_Adaboost model and kNN model to predict financial distress with the same datum, through comparing the predictions of three models, this paper find that SVM is significantly better than others.In summary, it has significant influence to the theory research and real economics by using SVM model to quantify and warn risk.