摘要： 将主成分分析(PCA)和支撑向量机(SVM)结合，提出了一种分行业、适用小样本空间的财务预警模型：PCA-SVM 模型。以传统财务指标为基础，通过主成分分析，简化输入向量，并利用SVM 作为判别企业状态的工具。该文利用了SVM 提取优秀企业各比率之间的内在相关知识，作为评判企业状态的依据，克服了以往在区分企业状态方法上线性判别的局限性，也克服了小样本条件下BP 网络推广能力不强的缺陷。
关键词:
主成分分析；支撑向量机；上市公司财务；企业危机预测
Abstract: This paper proposes a new method: PCA-SVM model which is composed of PCA and SVM to classify the state of enterprise through finance data. The model is fit for company in one special trade. It uses PCA to predigest the input vector and uses SVM to judge the statement of the company. It overcomes the localization of linearity distinguish in old models and it needs less samples to train the net than BP.
Key words:
PCA; SVM; Finance data of listed companies; Early warning of enterprise
李 贺，冯天瑾，丁香乾，张红兰. 企业财务预警 PCA-SVM 模型研究[J]. 计算机工程, 2006, 32(9): 233-235,238.
LI He, FENG Tianjin, DING Xiangqian, ZHANG Honglan. PCA-SVM Models for Early Warning of Enterprises Through Finance Data[J]. Computer Engineering, 2006, 32(9): 233-235,238.