基于复杂网络的上证指数序列分析

发布时间：2018-03-26 11:46

本文选题：复杂网络　切入点：粗粒化　出处：《华中科技大学》2012年硕士论文

【摘要】：研究股票市场规律是一门热门的课题，一方面是因为投资者期望能够找到一定的规律便于投资，另外一方面是因为学者们利用股票市场丰富的数据资源进行理论上的验证，但是一直以来大多数研究都是基于股票之间的关系，而本文从复杂网络的观点出发对上证指数波动进行探索。本文利用粗粒化方法，将上证指数2005年9月—2012年3月逐日收盘价序列转化为由5个字符{R,r,e,D,d}构成的上证指数符号序列。把连续的两个符号组合成一种波动模态，作为网络的节点（也就是连续三天的上证指数波动组合），然后按照时间顺序连边，构成一个有向加权上证指数波动网络。进而计算网络的度与度的分布，，最短路径长度，聚类系数，中介中心度等动力学统计量，研究蕴含在网络中的拓扑结构。另外结合数据挖掘中的聚类分析，将节点进行聚类，合并节点重建网络，深入挖掘上证指数波动过程中的有用信息。结果表明，上证指数变化具有复杂性，并且具有类混沌特征，不是完全随机的；并且在网络中，小幅波动模态节点中介中心度高，具有很重要的位置，这是与市场规律相符的，在实际的股票市场中，无论是牛市，熊市还是横盘，都离不开小幅涨或跌的调整。另外其他含有大幅变化的组合模态之间的转换可以为投资者提供有用的信息。
[Abstract]:It is a hot topic to study the law of stock market, on the one hand, because investors expect to find some rules to facilitate investment, and on the other hand, because scholars make use of the abundant data resources of stock market for theoretical verification. However, most studies have been based on the relationship between stocks, and this paper explores the volatility of Shanghai stock index from the point of view of complex network. In this paper, by using coarse granulation method, the daily closing price sequence of Shanghai Stock Exchange Index from September 2005 to March 2012 is transformed into a symbol sequence of Shanghai Stock Exchange Index composed of five characters {rrrrrrre / DU _ d}. The two consecutive symbols are combined into a fluctuating mode. As the node of the network (that is, the Shanghai Stock Exchange Index volatility combination for three consecutive days, and then connecting the edges in time order to form a directed weighted Shanghai Stock Exchange Index volatility Network, the distribution of the degree and degree of the network and the shortest path length are calculated. The topological structure contained in the network is studied by using the dynamic statistics such as clustering coefficient and intermediary centrality. In addition, the nodes are clustered and the nodes are merged to reconstruct the network by combining the clustering analysis in data mining. Deeply excavate the useful information in the fluctuation process of Shanghai stock index. The results show that the variation of Shanghai stock index is complex and chaotic, which is not completely random, and in the network, the intermediate center of the small fluctuation mode node is high and has a very important position. This is consistent with the laws of the market, in the actual stock market, whether it's a bull market, a bear market or a horizontal market, It's all about adjusting up or down slightly, and other conversions that contain significant variations in combination modes can provide useful information for investors.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F830.91;O157.5;F224

【参考文献】