| 混合HOGA-SVM财务风险预警模型实证研究 |
| |
| 引用本文： | 丁德臣.混合HOGA-SVM财务风险预警模型实证研究[J].管理工程学报,2011,25(2). |
| |
| 作者姓名： | 丁德臣 |
| |
| 作者单位： | 山东财政学院,山东,济南,250014 |
| |
| 基金项目： | 国际自然科学基金资助项目 |
| |
| 摘 要： | 目前涉及遗传算法与支持向量机相结合的预测模型中,遗传算法基本上采用的是标准算法.但是在对全局函数的优化中,一般的遗传算法容易陷入局部最优,从而降低遗传算法收敛速度和搜索精度,进而影响财务风险预警模型的精度与速度.基于此,提出了基于混合全局优化正交遗传算法(HOGA)和支持向量机(SVM)的财务风险预警模型(HOGA-SVM),通过使用混合全局优化正交遗传算法连同支持向量机来改进支持向量机进行财务风险预警的效果.结果显示,提出的模型不仅提高了财务风险预警的准确率和速度,而且模型的两类分类错误率(尤其是第一类分类错误率)相对其他模型也有了明显下降.未来的工作可以把模型的应用扩大到多分类的财务风险预警问题中.
|
| 关 键 词： | 财务风险预警 混合全局优化正交遗传算法 支持向量机 |
A Financial Risk Premonition Model Based on the Hybrid Orthogonal Genetic Algorithm for Global Optimization and Support Vector Machine |
| |
| DING De-chen.A Financial Risk Premonition Model Based on the Hybrid Orthogonal Genetic Algorithm for Global Optimization and Support Vector Machine[J].Journal of Industrial Engineering and Engineering Management,2011,25(2). |
| |
| Authors: | DING De-chen |
| |
| Institution: | DING De-chen(Shandong Financial University,Jinan 250014,China) |
| |
| Abstract: | Financial risk premonition can exert a significant influence on a company's survival and growth.Many financial risk premonition models generally fall into two categories:the traditional statistics model and the AI model.A hybrid model that incorporates hybrid genetic algorithms and support vector machines(SVM) is becoming an important financial risk promotion model.The simple genetic algorithm is frequently used in the hybrid financial risk premonition model.However,convergence rate has a weak performance i... |
| |
| Keywords: | Financial risk premonition hybrid orthogonal genetic algorithm for global optimization support vector machine |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
