摘要
以我国上市公司为研究对象,从深沪两市中选取了2002年新增ST公司41家和非ST公司41家为样本,使用SPSS软件,引入现金流量指标,从财务状况恶化前三年内两类公司20个指标中选取7个指标作为预测变量,分别运用多元线性回归和Logistic回归分析,建立并比较两种财务风险预测模型.结果表明:1)由于引入了以往研究中所没有的现金流量指标,使得模型在财务状况恶化前一年的误判率仅为6.10％;2)两种模型中,Logistic模型的预测准确率较高.
Based on china's listed companies' characteristics and risk management theories, using statistics and SPSS, the multivariate models for predicting financial risk of listed companies-LPM and Logistic Regression are studied and established. As cash flow ratios selected variable, the model is more precise in predicting financial risk with lowest error.
出处
《河北工业大学学报》
CAS
2003年第5期66-72,共7页
Journal of Hebei University of Technology
基金
河北省教育厅自然科学基金资助项目(2003302)
关键词
上市公司
财务风险
多元回归分析
LPM模型
LOGISTIC模型
listed companies
financial risks
multivariate regression analysis
LPM (Linear Probability Model)
logistic model
