基于混合两阶段模型的上市公司信用风险评价

Listed Companies′ Credit Risk Evaluation Based on Integration Model

利用支持向量机(SVM)-Logistic回归的混合两阶段模型来对上市公司信用风险进行评价。通过Logistic回归分析来对SVM的输出结果进行修正,降低了传统SVM方法的经验风险,提高了分类准确率。对SVM-Logistic回归模型、SVM和神经网络-Logistic回归模型进行实证比较,结果表明,支持向量机-Logistic回归模型的总判别准确率高于其他判别模型。

This paper uses the Support Vector Machine （SVM） and Logistic Regression for cnrpnrate financial risk evaluation. This can decrease the empirical risk of SVM outputs by interpreting and modifying the outputs of the SVM classifiers according to the result of Logistic Regression analysis so that it improves the accuracy of classifier rate. Compared with the models of SVM and NN - Logistic , the result shows that tire integrated binary discriminaut rule provcs marc accurate classifier rate than that of other models.