基于PSO的个人信用评估组合预测模型

将组合预测模型用于个人信用评估，在两种单一统计模型的基础上，利用粒子群算法（PSO）求解组合模型的权重，并通过粒子适应度函数的设置来控制第二类误判的发生，构建了基于PSO的组合预测模型。应用结果表明，基于PSO的组合预测模型的分类精度高于单一统计模型，并且有效降低了第二类误判率，对于商业银行控制信用风险具有更好的适用性。

Combining Forecast Model Based on PSO for Personal Credit Scoring

This paper used combining forecast model for personal credit scoring. Based on two statistical models, this paper constructed a combining forecast model using particle swarm optimization (PSO) to search the combining model's weights and using particle's fitness function to control the type II error rate. The application results indicate that the combining forecast gets higher accuracy with lower type II error rate on training samples and testing samples. The combining forecast model based on PSO presents mo...