上市公司财务状况评价的统计和神经网络方法  

Financial evaluation of listed companies based on ANN and statistic technology

作　　者：张维[1] 李洪良[1] 熊熊[1] 

机构地区：[1]天津大学管理学院

出　　处：《天津理工学院学报》2004年第4期33-36,共4页Journal of Tianjin Institute of Technology

基　　金：天津市自然科学基金资助项目(023600411)

摘　　要：根据上市公司的财务报表对上市公司的财务状况进行综合评价对投资者的投资和监管部门的监管活动具有重要的指导意义.本文利用主成分分析方法对产生于财务报表的各种财务比率指标进行优化处理,形成新的主成分指标,并以主成分作为输入变量,使用竞争学习网络进行聚类分析.最后用广泛采用的综合评分方法得出每一类企业的财务特征,清晰直观的表达了企业的财务状况.According to the financial statements of the listed companies, we can give a comprehensive financial evaluation of the companies to the investors and the supervesory administrations. It plays an important role for investors' decision and supervisors′ regulation. In this research, we use principal-component Analysis to process the data from the financial statements and get some optimal indicators named principle components. They are regarded as the input variables of the proposed competitive learning network, and we get a clustering result by using this ANN model. At last, we grade all the classes from the clustering result and get the financial features of each class of companies. We also use graphs to illuminate the results clearly and intuitively.

关 键 词：财务评价 神经网络 主成分分析 

分 类 号：F830.91[经济管理—金融学]