An Application of Complez Networks: Researches on Stocks Strong Correlation
In order to explore correlation characteristics among stocks, this article models complex network for 101 stocks in real estate sector of China Stock Market. In the model, nodes are stocks, and weighted edges are correlation coefficients of logarithm stocks returns in recently 17 years. We compute the topology parameters, and find that the network is free-scale, nodes degree distribution P(s)～|s|-δ with δ∈(0.8,1.6)for different threshold of correlation coefficients. Average cluster coefficient is 0.53. We also measure centralities of networks, seek for subgroups, and divide network into partitions. We find node 000592 and 601588 with high centrality, which influence entirety network greatly. The research shows that the method of complex network is effectual for revealing the emergence in stock markets.
Stock Market Network Topology Correlation Coefficient
LAN Wangsen ZHANG Suodi
Department of Mathematics, Xinzhou Teachers University, P.R.China, 034000 College of Management Science and Engineering, Shanxi University of Finance and Economics, P.R.China
国际会议
2009 International Institute of Applied Statistics Studies(2009 国际应用统计学术研讨会)
青岛
英文
1-5
2009-07-25（万方平台首次上网日期，不代表论文的发表时间）