Z-score模型最优分割点的确定方法比较——基于违约风险预测能力的分析

Predicting Default Risk of Z-score Model:The Weighted Average Method VS. The ZETAc Model Approach to Determine Optimal Cut-off Score

Z-score模型在对企业进行财务困境和违约风险判别方面具有重要的应用价值,最优分割点的确定方法对于提高模型的违约风险判别能力至关重要。本文以医药生物行业上市公司为样本,运用Fisher逐步判别法从15类财务比率中筛选出判别能力较强的7个指标构建了Z-score模型,并尝试采用加权平均法和考虑先验概率及误判成本的ZETAc模型法分别确定最优分割点。研究发现,ZETAc模型法预测企业违约风险的能力明显优于加权平均法。

The Z-score model has important application value in the aspects of predicting bankruptcy and default risk of enterprises. Methods to determine the optimal cut-off score are crucial to improve the model＇s default risk predicting ability. This paper uses Fisher Stepwise discriminant analysis to select 7 distinguish ratios out of 15 financial ratios with sample of bio- medicine industry listing corporation, and builds the Z-Score model. Then the default risk prediction results using ZETAc cut-off score is compared to that using weighted average cut-off score. The result shows that the prediction using ZETAc cut-off score is more accurate, so its forecasting ability of default risk is superior to the weighted average method.