摘要
为有效解决传统模型存在的小样本、高维数、非线性和局部极小点等问题,基于支持向量机方法建立了一种新的企业财务危机预警模型。该模型以财务危机预警指标体系为基础,企业财务危机与否的实际结果为学习样本,采用交叉验证和“格搜索”方法进行训练、验证,确定出最优分类函数。实例分析结果表明,该模型简单、有效、可行,为企业财务的动态预警提供了新的途径。
In order to solve the problem of small ,sample, high dimension, nonlinear and local minimizing of the traditional models perfectly, a novel model for pre-warning of enterprise financial crisis was established based on support vector machine. On the basis of the pre-warning index system, the actual results of enterprises＇ financial crisis were used learning samples to train and test the learning machine through cross-validation and grid-seareh, and an optimal classifier could be obtained, which was the target model. An example testified the simplicity, efficiency and feasibility of the novel model, which supplies a new approach to dynamic pre-warning of enterprise financial crisis.
出处
《中国石油大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2006年第4期132-136,共5页
Journal of China University of Petroleum(Edition of Natural Science)
基金
山东省自然科学基金项目(Y2003H01)
关键词
财务危机
预警
支持向量机
交叉验证
格搜索
financial crisis
pre-warning
suplmrt vector machine
cross-validation
grid-search
作者简介
张在旭（1959-），男（汉族），山东临朐人，教授，博士,博士生导师，从事油气工程管理、系统工程方面的研究
