菏泽学院学报

企业财务危机不仅影响自身的生存与发展,同时也使债权人及投资者的利益受到侵害,为此企业财务管理风险预测是不可或缺的,针对这一问题提出基于BP神经网络的企业财务管理风险预测实证分析.参考现有的研究成果,确定管理指标,将样本数据归一化处理,借助CSMAR数据库,补齐所缺失数据,利用BP神经网络,筛选出合适的样本,借助Matlab等统计分析软件,将Logistic回归预测结果及主成分分析预测结果,与应用BP神经网络的企业财务管理风险预测结果相对照,对照结果表明,BP神经网络的应用使现金流与企业财务安全的联系更加敏感,使预测结果更加准确真实,具有很强的操作性和实用价值.

Enterprise financial crisis not only affects its own survival and development,but also infringes on the interests of creditors and investors,therefore,the enterprise financial management risk prediction is indispensable.This paper proposes an empirical analysis of enterprise financial management risk prediction based on BP neural network.Referring to the existing research results,the management indicators are determined,the sample data are normalized,and the missing data are supplemented with the help of CSMAR database.The suitable samples are screened out by using BP neural network.With the help of statistical analysis software such as MATLAB,the results of logistic regression and principal component analysis are compared with the results of enterprise financial management risk prediction using BP neural network,showing that the application of BP neural network makes the relationship between cash flow and enterprise financial security more sensitive,makes the prediction results more accurate and real,and has strong operability and practical value.