摘要
公司财务困境受到决策者、市场、经济和政策多种因素影响,针对传统预测方法预测精度低缺陷,提出了一种贝叶斯判别分析的财务困境预测方法。首先选用了反映公司财务状况的24个指标作判别因子,建立了公司财务困境的贝叶斯判别分析模型,然后采用85个上市公司的实际数据作为学习样本建立贝叶斯判别函数,以交差确认估计法对判别准则进行评价,以验证模型的有效性,最后利用判别函数对5个待评价公司进行预测,得到判别函数值,进行仿真。结果表明,采用贝叶斯判别分析模型提高了公司财务困境的预测精度,是一种有效的财务困境预测方法。
Listed companies' financial distress prediction analysis is a complex systematic project,the self learning and adjustment ability of traditional forecasting methods are poor,thereby affecting the forecasting results.Using Bayes discriminant analysis theory and combining the current situation of listed company,24 indicators reflecting the company's financial situation were selected as the financial distress discriminant factors to build the listed companies' financial distress Bayesian discriminant analysis model.Using the actual data from 85 listed companies as learning samples,a Bayesian discriminant function was established to cross validation estimate method criterion for evaluation.In order to verify the validity of the model,finally discriminant function was used to forecast five companies and acquire discriminant function value.The results show that the listed companies Bayesian discriminant model has higher prediction precision,and is an effective method for the prediction of financial distress.
出处
《计算机仿真》
CSCD
北大核心
2012年第4期383-386,共4页
Computer Simulation
关键词
财务困境
预测
贝叶斯判别分析
Financial distress
Prediction
Bayes discriminant analysis