摘要
利用Logit回归模型和KMV模型分别对选取的29家我国上市公司样本的信用风险进行实证分析和比较。结果表明,两个模型基本上都能反映上市公司的信用状况,虽然KMV模型对非违约样本组的预测结果要优于Logit模型,但是在违约公司样本组的预测上,Logit模型要远比KMV模型准确。从总体效果来看,Logit模型的预测准确性要高于KMV模型。
The credit risk for the chosen twenty-nine listed companies in China was analyzed with logit regression model and KMV model, respectively. Results show that the two models can reflect the status of credit risk of listed companies basically. Although the prediction of KMV model on non-default group is better than that of the logit model, logit model is much better than KMV model on the the prediction of default groups. Overall, the prediction efficiency of logit model is higher than that nf KMV model.
出处
《青岛大学学报（自然科学版）》
CAS
2014年第2期90-95,共6页
Journal of Qingdao University(Natural Science Edition)
