【作者】 刘彦文；

【导师】 栾庆伟； 李延喜；

【作者基本信息】 大连理工大学 ， 技术经济与管理， 2009， 博士

【摘要】 竞争激烈的市场经济孕育发展机遇的同时,也暗藏着无尽的风险和危机。对于上市公司而言,因财务危机沦为“ST”甚至被迫退市的情况愈演愈烈。公司陷入财务危机,不仅危机其自身的生存和发展,也给投资者、债权人带来巨大的损失。因此,构建彼此不相关而且信息涵盖量大的财务危机预警指标体系,以及预测能力高的财务危机预警模型,对于证券市场的投资者和公司管理层而言,无疑能起到稳定证券市场、稳定国民经济乃至稳定社会发展的重要作用。本文在实证研究部分分别用Logistic预警模型、神经网络预警模型和遗传改进神经网络预警模型进行了预警模型的构建与检验。论文共分为五部分。第一部分介绍了上市公司财务预警的研究背景和研究意义,以及国内外研究现状;第二章介绍了企业财务危机的研究理论,分别从企业及其发展、企业财务危机的内涵和企业财务危机形成的原因进行了概述;第三部分主要对企业财务危机预警的相关理论进行了阐述,介绍了企业财务危机预警的理论依据,并对企业财务危机定性和定量预警的研究方法进行了回顾;第四部分是研究设计,主要介绍样本和预警变量的选取,并运用主成分分析提取主成分;第五部分是在分别用Logistic预警模型,BP神经网络预警模型和遗传优化神经网络预警模型进行了预警模型的构建与检验,并利用不同预测模型建模方法的实证结果进行比较;最后本文提出研究结论,并探讨了本研究的局限性和有关研究领域未来的扩展方向。本文的特色及创新之处在于:一是指标体系中引入了盈余管理、利润操纵程度指标和非财务信息,使预警指标体系更加完整。二是将遗传算法和神经网络相结合,构建了优化的财务预警模型。三是采用新的样本配对方法,增强了模型的适用性。更多还原

【Abstract】 While the dog-eat-dog market economy is pregnant with the opportunity for development,there are countless risk and crisis bided in it.As to listed company,the situation that it is titled with "ST" plate and even forced to quit listing because of financial crisis grows more and more seriously.It is not only the threat to existence and development of the company failing into the financial crisis,but also brings enormous loss to investors and creditors.So to establish a financial crisis early warning index system which is disrelated and covered amount of information,and build up a financial crisis early warning model,has become the important aspect of stabilifying the development of stock market,national economy,and society.In the empirical research part of this paper,we have separately used Logistic Early Warning Model,Genetic Algorithm and Neural Network to set up and test the early warning model.There are five parts in this paper.Firstly,background of research of financial risk and research significance is introduced,and the statue of research over the world is further introduced.Secondly,we have introduced the research theory of business financial crisis,and summarized the development of enterprises,meaning of corporate financial crisis and the formation of enterprise financial crisis.Thirdly,related theory of financial crisis early-waming is expatiated,and qualitative and quantitative methods of financial crisis early-warning are respectively reviewed.Fourthly,we selected the variables and samples to design the research,and extracted principal component factors.The fifth part builds up the financial crisis early warning models by Logistic Early Warning Model,BP Neural Network and GA-BP Neural Network after extracting principal component factors.And the empirical results are compared to get a conclusion.At last,we summarize the research conclusion and the shortcomings of our study and suggestments for further study.The main characteristic and innovation of this paper lie in two aspects.Firstly,the index system is more integrate for including non-financial indexes.Secondly,we build up a financial crisis early warning model,which combines Genetic Algorithm and BP Neural Network.Thirdly,we apply new matching method to select sample,and the applicability of the model increase evidently.更多还原