在对传统信用风险等级评价模型总结的基础上,提出了一种基于扩展粗糙集理论的评价方法.该方法首先将信用风险等级评价抽象成一个多准则分类问题,然后使用DRSA法直接从历史数据中提取决策规则,所得到的规则不仅可以用于企业的信用等级评价,还可以反映各指标与信用等级之间的关系.通过一组上市公司的实际数据验证了该方法的有效性.
Based on the overview of traditional credit risk evaluation models, a new approach using the extended rough theory is presented in this paper. This method first considers credit risk analysis as a multiple criteria sorting problem, and then applies DRSA to abstract decision rules from a number of history data. These rules can not only be used in the credit risk evaluation of firms, but also reflect the relationship between evaluation indexes and credit grades. At last, a practical example is used to illustrate the validity of this approach.