计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (20): 13-16.DOI: 10.3778/j.issn.1002-8331.2008.20.005

• 博士论坛 • 上一篇    下一篇

交叉熵算法在企业违约风险评估中的应用研究

周 泓,邱 月   

  1. 北京航空航天大学 经济管理学院,北京100083
  • 收稿日期:2008-03-10 修回日期:2008-04-18 出版日期:2008-07-11 发布日期:2008-07-11
  • 通讯作者: 周 泓

Application research on corporate default risk assessment based on cross-entropy algorithm

ZHOU Hong,QIU Yue   

  1. School of Economics and Management,Beihang University,Beijing 100083,China
  • Received:2008-03-10 Revised:2008-04-18 Online:2008-07-11 Published:2008-07-11
  • Contact: ZHOU Hong

摘要: 系统仿真是风险评价的一种重要手段,针对企业违约预测问题,提出了一种基于交叉熵算法的违约风险评判方法。采用公司未偿还贷款的概率作为衡量违约风险高低的标准,利用交叉熵方法构造企业违约风险识别模型及其算法,并由此估计出发生损失的概率。与传统的预测方法进行比较,结果表明该模型对违约风险具有很强的识别能力,预测精度高。

关键词: 交叉熵, 风险评估, 仿真, 违约风险

Abstract: System simulation is one of important tool for risk assessment.A new method is presented to deal with corporate default forecast problems based on cross-entropy algorithm.The failure probability of repaying loans of corporation is taken as the criterion to measure the level of default risk.The cross-entropy scheme is adopted to construct the model of default risk identification,based on which the loss probability can be assessed.Contrasted to traditional forecasting methods,the forecasting method has a strong capability to identify the default risk and high forecasting precision.

Key words: cross-entropy, risk assessment, simulation, default risk