摘要
将主成分分析(PCA)和最小二乘支持向量机(LSSVM)结合,提出了一种适用小样本空间的财务风险模型:PCA—LSSVM模型。以传统年度财务指标为基础,通过主成分分析,简化了输入变量,并利用LSSVM作为判别企业风险等级的工具。该模型可以在纺织行业中的上市公司选中1个或多个企业参与,避免了传统算法模型在解决财务风险预测的缺陷。算例结果表明了所提出模型能有效地提高预警方案的可行性,为财务风险的在线实施提供了方便。
This paper proposes a new method:PCA-LSSVM model which is composed of PCA and LSSVM to protect company from financial risk. The model is fit for company in the small sample space. It uses PCA to predigest the input vector and uses LSSVM to judge the statement of the company risk. With the model,one or more listed company could be selected in textile industry,and the defects in resolving early warning of The fnancial risk with traditional algorithm model could be omitted. The application case proved that the proposed method can improve the feasibility of the program in financial risk,and it is suitable for on-line financial risk control .
出处
《中国电子商务》
2010年第12期213-214,共2页
E-commerce in China
关键词
主成分分析
最小二乘支持向量机
纺织行业
财务预警
principal component a,nalysis
least squares support vector machine,textile industry
early warning analysis