节点文献
概率神经网络在财务预警实证中的应用
An Application of Probabilistic Neural Network to Warning for Corporate Financial Distress
【Author】 WU De-sheng, LIANG Liang(School of Business, University of Science and Technology of China, Hefei 230026, China)
【机构】 中国科学技术大学商学院;
【摘要】 建立切合企业实际的财务预警系统,具有降低企业经营风险、投资风险以及防范金融危机的积极作用。在充分考察财务困境研究领域现状的基础上,建立一套适合于我国企业的财务状况识别指标体系。本文将概率神经网络(PNN)应用在财务困境研究领域,建立了财务预警模型。结果表明,该预警模型具有较高的预测准确率和良好的操作性。其短期(一年期)和中期(三年期)预测准确率分别为87.5%和81.25%。
【Abstract】 To build up a warning system of an enterprise’ s financial risk using the estimated probability of financial distress is indispensable in reducing a firm’ s operation and investment risk and preventing national finance crisis. In view of the actuality of corporate financial distress prediction, we design some distinguished ratios. The probabilistic neural network (PNN) is proposed to predict financial distress of Chinese corporations. The demonstration’results account for the high accuracy and discriminant power of the PNN model. The short-term accuracy and medium-term accuracy are 87.5 % and 81.25% respectively.
【Key words】 financial distress; probabilistic neural network; prediction; model;
- 【会议录名称】 2003年中国管理科学学术会议论文集
- 【会议名称】2003年中国管理科学学术会议
- 【会议时间】2003
- 【分类号】F224
- 【主办单位】中国优选法统筹法与经济数学研究会