个人信用评估组合模型选择方案研究

The research on methods of personal credit scoring combined model selection based on optimized index system

为准确评估借款人的信用和合理控制商业银行风险,首先对个人信用评分模型的常用模型进行了总结和归纳,然后通过分析比较说明了不同模型的准确性.在"坏样本"的区分与结果判断上采用加权方式,提出修正算法来确定信用评分模型中的指标权重,满足不同银行数据多样化的需要,提高评分模型精度.

For precise estimation of borrowers＇ personal credit and reasonable risk management of commercial bank, main models as well as problems are pointed out first. Next, to solve these problems, a modified algorithm is designed to compute a series of weights for indexes to satisfy different needs in different banks with various data and finally improve the accuracy of the model.