摘要
利用2017-01-03—2018-12-31时段标普&500指数(美国)、富时马拉西亚指数(马来西亚)、瑞士SMI指数(瑞士)、比利时20指数(比利时)、中国上证指数(中国)、意大利富时指数(意大利)6个国家的股票每日收盘价数据,基于有限穿越可视图理论构建了股票数据时间序列网络,通过计算网络的度分布、聚类系数、介数等特征来分析其拓扑结构.结果表明这些网络具有小世界特性,度分布呈现幂律特性等特点,也揭示了股票市场的部分特征.最后,探讨了基于有限穿越可视图的股价预测.
Based on the daily closing price data of S&P 500 index(USA),FTSE Malaysia index(Malaysia),Swiss SMI index(Switzerland),Belgium 20 index(Belgium),Shanghai stock index of China(China)and FTSE index of Italy(Italy)from January 3,2017 to December 31,2018,the stock data time series were constructed by the limited penetrable visibility graph theory.The topological structure was analyzed by calculating the degree distribution,clustering coefficient and betweenness of the network.The results show that these networks have the characteristics of small world,power-law of degree distribution and some characteristics of stock market.Finally,the stock price forecast based on the limited penetrable visibility graph is discussed.
作者
祝嘉
向长城
ZHU Jia;XIANG Changcheng(School of Mathematics and Statistics,Hubei Minzu University,Enshi 445000,China)
出处
《湖北民族大学学报（自然科学版）》
CAS
2020年第3期283-289,共7页
Journal of Hubei Minzu University:Natural Science Edition
基金
国家自然科学基金项目(61763009).
关键词
时间序列
股票数据
有限穿越可视图
复杂网络
time series
stock data
limited penetrable visibility graph
complex network
