首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于BP神经网络的大非减持影响因素实证分析——以深交所上市公司为例
引用本文:曹国华,赵晰.基于BP神经网络的大非减持影响因素实证分析——以深交所上市公司为例[J].软科学,2010,24(5).
作者姓名:曹国华  赵晰
作者单位:重庆大学,经济与工商管理学院,重庆,400044
基金项目:国家社会科学基金资助项目,教育部新世纪人才 
摘    要:基于BP神经网络,以深交所上市公司大非股东为样本建立模型,利用遗传学算法对模型结构进行优化并通过神经网络贡献率分析法得出对大非减持贡献率较大的因素。对在指标变量几种典型取值状态下的关系曲线进行深入研究后认为,大非减持并非全然表示企业投资价值下降,普通投资者亦可从中获取积极投资信息;对监管者而言,充分保证大股东流通权以维持其对国家政策的信心,立足贡献率较大的因素调控大非减持比单纯进行限制更能维护市场稳健,实现监管者、普通投资者与大非三方共赢。

关 键 词:BP神经网络  遗传算法  大非减持  贡献率

An Empirical Analysis of the Subtraction Influencing Factors of the Sale of Large Non-tradable Share Based on BP Neural Network——Taking Shenzhen Stock Market as an Example
CAO Guo-hua,ZHAO Xi.An Empirical Analysis of the Subtraction Influencing Factors of the Sale of Large Non-tradable Share Based on BP Neural Network——Taking Shenzhen Stock Market as an Example[J].Soft Science,2010,24(5).
Authors:CAO Guo-hua  ZHAO Xi
Abstract:Based on the BP neural network,the paper sets up the model with the sample of large non-tradable shareholders in Shenzhen stock market and optimizes the model by genetic algorithm,and then gets factors of the more contribution for output by neural network contribution rate analysis.Through deeply studying the related curves under the different index variables,the results of empirical research show that the sale of large non-tradable share is not all the representative of the declining investment value of the enterprise,and ordinary investors can get positive investment signals from it.In addition,regulators should completely guarantee the circulation right of large shareholders to maintain its confidence in the national policy,and control the sale of large non-tradable share based on those factors with higher contribution rate,which is more effective to maintain the market stability than limiting it only.It brings a multilateral beneficiary situation to regulators,investors and large non-tradable share.
Keywords:BP neural network  genetic algorithm  sale of large non-tradable share  contribution rate
本文献已被 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号