论文部分内容阅读
目前企业违约预测模型和现实情况存在一定差距,表现在:1)违约公司与正常公司样本数比例与实际情况严重不符;2)已有的研究极少考虑误判损失;3)鲜少提及信用等级,行业,规模,地区等定性指标对违约预测的影响.针对以上问题,建立了一个考虑误判损失的违约预测Logistic模型,摒弃以往配对原则,采用全样本分析,将地区、规模、行业作为定性指标和29个财务比率指标代入Logistic逐步回归后,最后得到一个违约判别模型.
At present, there is a certain gap between the default forecasting model of enterprises and the actual situation, which is manifested in the following aspects: 1) the ratio between the number of default companies and normal companies is seriously inconsistent with the actual situation; 2) the existing research seldom considers the misjudgment of losses; and 3) Credit rating, industry, size, region and other qualitative indicators on the impact of default.According to the above problems, this paper establishes a default forecast Logistic model considering miscarriage of justice, abandoning the principle of matching in the past, using the full sample analysis, the region, size, industry As a qualitative indicator and 29 financial ratios into Logistic regression step by step, finally get a default discriminant model.