The Study Of High-tech Enterprises Financial Distress Prediction Based On SVM Ensemble

The high-tech enterprise and industry development is the strategic high groundof today’s world economy, science and technology competition. High-techenterprises are established on the basis of scientific and technological innovation.High-tech enterprises invest a lot and yield huge profit. However, marketcompetition is fierce, the uncertainty of economic policy and the acceleration oftechnological upgrading directly or indirectly lead to financial risks. Building amodel to predict financial crisis can guard against financial risks and protectfinancial health.Financial crisis is the explosion of financial risks. Building a prediction systemcan help monitor the productive activities, inform potential risks and make thedecision. The article studies the influncial factors and transmission mechanism offinancial risks. Finally we choose18prediction indicators.The third chapter builds SVM Ensemble model. SVM method has theadvantage to classify small sample size, high dimensional, non-linear characteristicsof the data, which displays a unique superiority and a good prospect, thus helping topredict financial trouble of high-tech enterprises. By reselecting the training set,bagging integration increases classification accuracy and improves generalizationability. Existing weak classifier fusion techniques only consider the credibility of theclassification accuracy, but not take full advantage of the sample data. To solve theseproblems, we propose evidence-based fuzzy density calculation method. To measurethe credibility of the SVM output, we consider3aspects: the output credibility ofsub-classifiers; the credibility of sub-classifiers; the differences betweensub-classifiers.The fourth chapter carries out an empirical test to prove the accuracy of theproposed model and the efficient use when it’s put into practice. On the basis ofprevious research, we choose the data of2007ST high-tech enterprises as theresearch object to carry out a empirical study. By comparing a single SVM modeland SVM Ensemble model, the results show that the proposed model predicts ahigher accuracy rate than a single SVM model, and the result confirms thefeasibility and effectiveness of the model, provides a method for the listed high-techenterprises to establish the reliability of financial distress prediction system.