A Research On Financial Distress Prediction In China's Pharmaceutical Companies Based On SVM

The research on enterprise financial distress prediction has started for a long term, and its importance and application has been widely accepted.Based on the domestic and international literatures,the research of enterprise financial distress prediction stayed at the domains of building a common and general research.However,with the rapid economy development,the difference of companies in different industries becomes gradually evident.Thus,constructing the index system and model of financial distress prediction for every industry has been a research issue at present.Pharmaceutical industry was considered as the "eternal sunrise industry",but in fact,the pharmaceutical manufacturing is a typical high investment,high-risk,highreturn industry,the risk of this industry is higher than the average,especially in China. At present,with the globally financial crisis,it has increase the possibility of pharmaceutical industry to be greatly affected by the crisis.Thus,implement of financial distress prediction system for pharmaceutical industry will minimize and avoid the risk of investors,creditor,and other stakeholders.Based on the concepts of financial distress prediction and review of related literature,this research paper develops a model of financial distress prediction for pharmaceutical industry with the data gathered from pharmaceutical industry in China. After introducing the concept and content of financial distress prediction and the review of literatures about related studies on financial distress prediction,the paper utilize the financial data of sample companies,implement prediction index system by a combination of quantitative and qualitative methods with a view to characteristics of the industry.And then,the paper applies the two structural kinds of prediction model on SVM and test the prediction accuracy of SVM model through an actual experiment.The experiment shows that,the financial distress prediction model based on SVM has an excellent result of which the prediction accuracy is 91.67%(11/12),for the pharmaceutical industry of China with less trained samples of a certain industry. Therefore,SVM deserves being applied widely in financial distress prediction in a certain industry.