节点文献

中国上市公司财务危机阶段性特征判别研究

【作者】 罗叶

【导师】 杨作廪;

【作者基本信息】 西南财经大学 , 统计学, 2004, 硕士

【摘要】 (一)任何企业都有自身的生长周期,从诞生开始就面对死亡,而现代社会日益激烈的竞争似乎加剧了这一过程。从国内的情况看,WTO和公司制改革的双重压力,使企业的生存变得更加艰难。企业经营管理者、政府和投资人的多方需要正为企业危机预测这一课题带来前所未有的市场。企业风险是不确定因素导致企业收益偏离预期。企业风险的表现形式多种多样,而任何形式的风险最终都会表现为企业的财务波动,并会在企业的财务报表上有所反映。从这个意义上讲,财务风险是企业的本质风险,本次研究也将从财务角度考察企业风险。风险可以是超出预期的有利事件,也可能是超出预期的不利事件。企业各方利益群体对意外损失的关注远大于他们对意外收益的关注。当财务风险表现为超出预期的意外损失时,就成为财务危机。从实证研究的角度,国内的研究人员普遍选择被特别处理(ST)作为财务危机的标志。这一思路实质上借鉴了国外研究将财务危机定义为破产的做法,简化了危机的内涵,赋予样本企业以明显的标识。同时,国内的研究普遍将非ST企业作为财务安全企业样本与ST企业配对,笔者认为这是对危机内涵的模糊。财务危机贯穿企业生命周期的始终,并没有任何企业处于绝对安全阶段。ST意味着企业的财务危机达到某种严重的程度,而非ST企业事实上也面临不同程度的财务危机,将其作为财务安全企业样本不符合危机的定义。本文拟对此进行改进,将ST企业定义为财务严重危机的样本企业,并配合财务轻度危机和财务稳健企业样本进行研究。(二)从Beaver的单变量模型开始,研究人员不断尝试变量、样本和<WP=4>模型方面的改进,提高财务危机实证预测技术,并在这一领域取得了若干突破性研究成果。从变量的角度看,研究人员先后尝试了加入现金流量指标、审计意见、产业相对比率、来自宏观和微观角度的非财务因素和增长率指标来提高模型的判别能力,拓展模型的适应性。从样本的角度看,研究人员发现并证实了样本选择偏见的存在,并尝试采用单一行业样本来消除这种对模型预测精度的高估。从模型的角度看,在Beaver的单变量模型和Altman的线性多元判别模型之后,bayes判别、logistic回归、逐步回归、离位分析、Probit模型、多变量距离分析、专家系统(ESRA)方法、神经网络方法也被引入这一领域,研究人员还尝试对各种预测方法进行比较,大大丰富了财务危机预测模型的内容。此外,在研究设计上,很多研究人员放弃了最初对样本企业的两分类判别,根据财务危机的特性设计了三分类甚至多分类样本分组模式,赋予财务危机更多层次的内涵。国内的研究开始于90年代末期,主要研究成果来自于借鉴国外模型并分析其在国内的适应性。从总体研究水平看,笔者认为存在若干不足,例如:财务危机概念混乱,认为企业不是面临财务危机即是处于财务安全状态,从而仅将其识别为危机与安全;在变量的选用上没有考虑国内不同于国外的特殊情况,例如没有考虑盈余管理和利润操纵造成的会计信息质量问题;变量筛选侧重于定性分析,仅有少数研究成果使用定量方法筛选变量,仍然存在筛选过程中变量不充分造成筛选不充分、对筛选结果不做任何定性分析直接用于建立判别模型的缺陷;在对企业财务状况的好坏究竟由财务指标在某一年度的水平还是其在某一时段的持续表现决定没有定论的情况下,对财务数据不经挑选直接使用单一年度截面数据进行判别;同时判别方法的比较也不够全面,未能区分在不同分类方式下的判别效果。本次研究拟在这些方面加以改进。 <WP=5>(三)本文希望通过实证分析得到两个方面的结论:了解企业财务状况更容易受那些财务指标的左右,更多地取决于企业财务比率指标的短期水平或长期累积表现;衡量使用实证方法识别企业财务质量阶段性特征和恶化程度的可能性,评价不同性质的方法在样本不同分组条件下的适应性和识别效果。为此,笔者设计使用逐步选择的方法,分别单一年度财务指标和四年平均财务指标筛选变量,配合相应的样本数据建立两分类fisher判别方程,根据判别效果的优劣确定变量和样本数据的最优组合,用于企业财务危机阶段性特征判别研究。在具体判别的过程中,为了识别企业财务危机的阶段性特征,本次研究将样本划分为财务稳健企业与财务危机企业,同时将财务危机企业进一步划分为财务轻度危机企业和财务严重危机企业。采用常规多元统计方法、非参数判别方法和径向基神经网络模型三个层次的判别方法进行判别。具体判别时过程分为直接判别和分步判别,即直接将样本单位识别为财务稳健、财务轻度危机和财务严重危机企业或是先将样本单位识别为财务稳健企业与财务危机企业,然后进一步将财务危机企业识别为财务轻度危机企业和财务严重危机企业,借此评价不同判别方法在样本两分类和样本三分类情况下的判别效果。(四)在现有财务危机识别研究的基础上,本次研究进一步探索了财务危机阶段性特征的识别的方法,并在以下方面有所发展:在以往研究仅对变量进行挑选的基础上,对样本数据进行了选择。对单一年度样本数据和四年平均样本数据分别组合不同的变量,并通过判别方程判别效果的优劣,获得了四年平?

【Abstract】 (一)There is a growth cycle in any enterprise, since born it faces the death, and competition seems to aggravate this course day by day. Seen from domestic situation, it makes the existence of enterprise becomes more difficult because of company reform and WTO. It brings the unprecedented market demand for financial distress predited by many enterprise administrators, government and investors.Enterprise’s risk is that uncertain factors lead to the fact that enterprise deviates from expectancy by income. The behavior of enterprise’s risk takes many forms, and any form will show as the financial affairs of enterprise fluctuation, and will reflect in the financial statement frequently. So the author thinks that the financial risk is the essential risk of an enterprise, and enterprise’s risk will be investigated from the financial angle in this thesis .The risk can be favorable incident or be unfavorable incident that goes beyond the anticipation. People always pay more attention to accidental loss than unexpected income. It becomes the financial crisis when the financial risk is shown as an accidental loss.From the angle of the empirical research, domestic researchers generally regard Special Treatment (ST) as the sign of the financial crisis. This train of thought has been drawn from the studies that defines the financial distress to be bankrupt, and has simplified the intension of the crisis, and give the sample obvious sign. Furthermore, domestic research regard non-ST enterprise as the safe enterprise sample and mix non-ST enterprise with ST enterprise generally. The author thinks this is fuzzy to the meaning of distress. Throughout, any enterprise is not in the definitely safe steps, and the financial distress will accompany the whole life of the <WP=8>enterprise. ST means that the financial distress of the enterprise reaches a certain serious degree, but non-ST enterprise faces the financial distress in varying degrees in fact, Regard it as definitely safe enterprise sample does not accord with the real mean of distress. This thesis plans to improve this, It is a sample enterprise of a serious distress of the financial affairs to define ST enterprise, and cooperate with the slight crisis and sane enterprise sample of financial affairs to study.(二)From single variable model of Beaver, researchers try to get improvement of the variable, sample and model constantly, improving the case study and predicting technology of financial distress, and has made several breakthroughs in research in this field .In order to improve the qualify of variables, researchers have tried to join the cash flow index, Audit suggestion, industry relative rate, non-financial factors from the macroscopic and micro perspectives and rate of increase index to the model. From the angle of sample, researchers have found and verified the existence of choosing- prejudice in sample, and try to adopt the single trade sample to dispels this question and predict the over-evaluating of precision to the models .From the angel of model, after the single variable model of Beaver and linear multivariate classify models of Altman, Bayes classify, logistic regression, stepwise regression, probit model, ESRA , and neural network methods are introduced into this field. Researchers still try to compare various kinds of prediction methods, and have enriched the financial content of predicting model greatly .In addition, in studying the design, a lot of researchers have given up classifying the sample to two groups and design to classify the sample into three or more groups according to characteristics of financial distress, and give the intension with more multi-level financial <WP=9>implication.Domestic financial distress research began with the latter part of the 1990s, most of the research results are borrowed from foreign model and the researchers analyse their adaptability at home. From the overall research standard, The author thinks there are several defects, such as: financial distress confused concept, and assumption that en

  • 【分类号】F275
  • 【被引频次】3
  • 【下载频次】456
节点文献中: 

本文链接的文献网络图示:

本文的引文网络