【作者】 白雪；

【导师】 刘晓春；

【作者基本信息】 北方工业大学 ， 数量经济学， 2006， 硕士

【摘要】 我国资本市场经历了十几年的探索发展,取得了很大的成绩,且其在国民经济中的地位在2004年得以高度确认。而上市公司作为资本市场的主体,各方面倍受关注。近年来,随着一些表面业绩良好的公司的纷纷破产,上市公司财务危机预警的研究成为越来越引起重视的一方面。各方对财务业绩、财务危机及破产的预测研究需求日益迫切,财务预警研究也不再仅仅是一个学术问题,更成为影响我国企业及资本市场发展的重要因素。 本文以我国上市公司作为研究对象,以因财务状况异常而被列为特别处理公司(ST公司)作为界定上市公司财务危机的标志,利用公开的财务报表数据,采用Logistic回归和BP神经网络寻找最佳的预警我国上市公司财务危机的模型和变量。文中的Logistic回归和BP神经网络分别使用SPSS和Matlab实现。研究结果表明,利用前一年年报的财务数据,本文所构建的Logistic回归模型和三层的BP神经网络能够较准确预警出在下一年进入ST板块的公司,而具有显著解释能力的财务指标为:EVA、总资产报酬率、主营业务利润/总资产、净资产、主营业务现金比率、成本费用利润率。更多还原

【Abstract】 Chinese capital market has been developing for more than ten years and has got prodigious achievement. Its status in domestic economy had been high affirmed in 2004. Recently, more and more corporations seeming good get into insolvent, then finance distress alter become an important field. All circles demand the study on finance fruit and finance distress. Finance distress is not only a learned problem, but has been an important factor to our corporations and capital market’s development.This thesis makes public corporations as the study objects, ST stock corporations because of exceptional finance as criterion to class finance distress. Logistic regression analysis and BP neural network are used to find the optimum finance distress alert model and variables with exoteric financial data. SPSS and Matlab are used to implement them. The results demonstrate that using the last year’s financial statement, this thesis’s models could forecast the ST stock corporations this year. EVA、 the assets reward ratio、 income from main operation、 assets、 net assets、 net cash flow from operating activities/sales- net income/cost of sale plus financial cost plus overhead cost plus marketing cost are distinct explanative financial index.更多还原