Support Vector Machine Model of Financial Early Warning

摘要：

The article is selected 66 listed companies of 13 industries of Shanghai and Shenzhen stock market as samples, and selected three-year consecutive financial statement data in the year of 2007~2009. We use support vector machine to set up a new financial early warning to predict the selected listed companies. In order to test the validity of SVM prediction, we compare it with BP Neural Network, which is more accepted in nowadays. We find that because of its small sample research and unlimited dimension, SVM model has more advantages and higher accuracy than BP Neural Network.

关键词： Financial early warning Support vector machine Empirical analysis

作者: Chen Hong Liu Jingshu

作者单位: School of Economics and Management, Zhongyuan University of Technology, Zhengzhou, China

会议类型: 国际会议

会议名称: 2010 International Conference on Circuit and Signal Processing(2010年电路与信号处理国际会议 ICCSP 2010)

会议地点: 上海

会议语种:英文

页码: 46-49

在线出版日期: 2010-12-25（万方平台首次上网日期，不代表论文的发表时间）