摘要
针对传统的Z-Score财务预警模型预警能力的不足,导致无法准确判定上市公司的财务风险状况,将SOA算法的良好寻优能力和Z-Score财务预警模型结合起来,提出一种改进的Z-Score财务预警模型,构建出SOA算法优化Z-Score财务预警模型的适应度函数。仿真对比发现,改进的Z-Score财务预警模型其平均识别率高达96.33%,远远高于SVM算法和AdaBoost算法的平均识别率,改进的算法极大地提升了Z-Score财务预警模型的预测能力,使其更具适应性。
The conventional Z-Score model financial early warning lacks predicative power,making it impossible to accurately determine the financial risk profile of listed companies.It needs to be further optimized to enhance its power.This article combines the optimization ability of SOA with Z-Score financial early warning model algorithms,and proposes an improved Z-Score financial early warning model to construct a SOA algorithm optimization fitness function for the new early warning model.Our simulation result show that the improved Z-Score financial early warning model increases the average recognition rate up to 96.33%,much higher than the average recognition rate of SVM algorithm and AdaBoost algorithm;and the improved algorithm greatly enhances the ability of the Z-Score Financial Early Warning Model by making it more adaptable.
出处
《财经理论与实践》
CSSCI
北大核心
2015年第2期66-70,139,共6页
The Theory and Practice of Finance and Economics
基金
教育部人文社会科学规划基金(14YJA790089)
关键词
Z-SCORE模型
人群搜索算法
寻优能力
数学模型
适应度
Z-score model
Seeker optimization algorithm
Optimization capabilities
Mathematical model
Fitness