基于STSA的股市网络结构实证分析

Network structure empirical analysis of the stock market based on STSA

网络分析法为分析股票市场的复杂性提供了有效的方法.文章采用符号时间序列分析法,使用符号编码序列所对应的频率,基于欧几里德空间,计算最小生成树节点之间的距离.同时,引入多维数据分析法,在收益序列的基础上加入股票交易量信息,建立最小生成树和分层树,从而体现出交易量作为股市价格趋势指向标所隐含的信息.最后,对沪深300指数进行了实证分析,系统分析了其网络结构及聚类特性.

Network Analysis Method offers an effective method for analyzing the complex stock market. In the present paper, adopting Symbolic Time Series Analysis （STSA） method, using the frequency of the coded symbolic series, combining with the concept of Euclidean Space, the distances among the nodes of minimal spanning tree are calculated. Meanwhile, introdu- cing multidimensional data analysis method, which means including the trading volume information, minimal spanning tree and hierarchical tree are constructed, and information implicated in trade volume - the tendency remark of price is embodied. Finally, empirical analysis of the component stocks of Hushen 300 index is made.