基于离散Hopfield网络的上市公司财务困境预警研究  

An Early Warning Research on Financial Distress of Listed Companies Based on Discrete Hopfield Network

作　　者：杨旸[1] 林辉[1] 

机构地区：[1]南京大学商学院,江苏南京210093

出　　处：《华东经济管理》2016年第12期156-162,共7页East China Economic Management

基　　金：国家自然科学基金面上项目(71271110);中国经济改革研究基金会资助项目(2016Z-002);2016年南京大学研究生跨学科科研创新基金资助课题

摘　　要：上市公司陷入财务困境直接影响投资者收益、管理者决策和股东利益,建立行之有效的财务困境评价模型已成为学术界关注的焦点。文章在阐述模型和构建指标体系的基础上,提出基于离散Hopfield网络模型开展上市公司财务困境评价的步骤,以2013-2015年沪深A股被实施ST的上市公司为实证样本,进行财务困境预警实证研究和稳健性检验。实证结果表明,本模型至少可以提前1年对ST上市公司的财务困境实现全面预警,可以提前6个季度对92%的ST上市公司进行有效预警,故适用于财务困境预警研究且具有稳健性。The financial distress of listed companies has a direct impact on the earnings of investors, decision-making of managers and interests of shareholders. The establishment of an effective evaluation model of financial distress has become the focus of acadenlic attention. On the basis of model elaboration and index construction, this paper puts forward the implementa- tion steps of financial distress evaluation based on discrete Hopfield network model, and then carries out financial distress fore- warning empirical research and robustness test on the samples of ST listed companies in Shanghai and Shenzhen A shares from 2013 to 2015. Empirical results show that this model can achieve a comprehensive forewarning on the ST listed companies＇ financial distress at least one year ahead, and can achieve effective forewarning on 92% ST listed companies six quarters ahead. The model is suitable for financial distress prediction research and has robustness.

关 键 词：上市公司 财务困境 离散HOPFIELD网络 预警研究 

分 类 号：F275[经济管理—企业管理]