基于复杂网络理论的股票指标关联性实证分析

Empirical Analysis of Relevance of Stock Indicators Based on Complex Network Theory

复杂网络理论是研究股票市场内在结构和功能的有力工具,股票指标的关联性已成为研究股票市场的一个新视角。基于复杂网络理论对深沪300指数(2011年7月28至2012年2月28)构建网络拓扑结构。利用DFA进行数据筛选,计算筛选后数据的股票指标的绝对相关系数。根据股票指标的绝对相关系数概率分布确定阈值,构建网络拓扑结构,计算股票指标的网络统计特征指标,分析收益率、成交量、市盈率的网络结构。研究结果表明,股票收益率和成交量指标的具有较强的关联性,具有小世界性质;市盈率指标具有较弱的关联性,不具有小世界性质,是随机网络且受外界影响较大、效率低下,风险比较大。

The complex network theory is a powerful tool to study the internal structure and function of the stock market. The research of the relevance of stock indicators has become a new perspective in the re- search of stock market. Traditional financial research using equity-linked network to dig intrinsic cluste- ring structure, lack of formulation of the relationship between the complex structures. This paper based on complex network theory to build the network topology on CSI 300 Index （July 28 2011-February 28 2012）, the relationship between the complex structures is analyzed, the existing empirical analysis methods of complex network are improved. The data is tiltered by DFA, and the absolute correlation coefficient of the stock index is calculated. According to the absolute correlation coefficients of stock index probability dis- tribution, the threshold is determined and build the network topology is build. Network statistics charac- teristic index of the stock is calculated, and the network structure of the rate of return, trading volume and price-earnings ratio are analyzed. The results show that the stock returns and trading volume indicators have strong correlation with the small-world character,while price-earnings ratio index has a weak correla- tion, and does not have the small-world character, and is a random network that subject to outside influ- ence with low efficiency and high-risk. Such new tools and empirical results are useful to study the rele- vance and empirical findings of stock indicators in China＇s financial market.