Research On Financial Markets Based On Complex Networks

Complex network theory is one of the important theories of complexity systems. It is mainly through the network in the form of modeling complex system. The complex networks are built taking the objects as nodes, with the correlation of the objects taken as boundaries. Then, the statistic characteristics of the complexity systems are analyzed utilizing the theories of the complex network.Financial market is a very complex system. It has a large number of investors and investment products. It also concludes a lot of investment behaviors of investors in the complex financial markets. It is very suitable for using the theory of complex networks. The complex networks are built taking the investment products as nodes, with the correlation of the investment products taken as boundaries. The financial market rules can be found through the research on statistic characteristics of the networks.This dissertation constructs complex networks with the price data of the international and domestic stock and futures markets. The global properties of the networks are analyzed utilizing the theories of the complex network. Then the theory and practice results are found. The main research works of this dissertation are as follows:(1) Research on International Stock and Futures Indexes Based on the Complex Network TheoryThe relativity and the overall characters of the international stocks are studied based on the complex network theory. The relation between the international stocks indexes and the economics is analyzed. The unweighted and weighted complex networks are built taking 61 international stocks indexes as nodes, with the correlation of the international stocks indexes'fluctuation taken as boundaries. Then, the topological structures of the networks are analyzed utilizing the theories of the complex network. The relativity and the overall characters of the futures indexes are studied based on the complex network theory. The relation between the futures indexes and the industries is analyzed. The unweighted and weighted complex networks are built taking 40 futures indexes in the Webstock as nodes, with the correlation of the futures indexes'fluctuation taken as boundaries. Then, the topological structures of the networks are analyzed utilizing the theories of the complex network. The impact of the futures indexes on the futures market reflects the impact of the futures varieties on the world economics.(2) Characteristic Analysis of Stock Short-term Risk Complex Network of the Shanghai Stock MarketIt simulates stock short-term risk with the VaR arrays of stocks. The undirected and unweighted stock short-term risk networks are built taking stocks in Shanghai stock market, with the correlation of the VaR arrays of stocks taken as boundaries. Then, the topological structure of the network is analyzed utilizing the theories and methodology of complex networks. Some suggestion can be provided for the actual investment operation.(3) Research on Mixing Patterns and Network Navigation of Chinese Stock MarketThe stocks are classified according to region, industry, market capitalization respectively. The mixing patterns are researched according these three classification criteria. Then, the network navigation is researched in accordance with the industry classification standard. The results are following:Shanghai market and Shenzhen main board show assortative mixing in accordance with the industry classification standard with high threshold. Chinese stock market shows a little disassortative mixing with low threshold, while it shows assortative mixing with threshold reaching a certain level. Network navigation phenomenon in the stock market reflects the interaction between the disparate stocks. There exists much network navigation in Chinese stock market.(4) Fractal Characteristic of the Chinese Stock Market Complex NetworkBy using the threshold method, it constructs the Chinese stock market complex network and analyzes the fractal characteristics of the network from the dimensions of time and space. Firstly, the static network is analyzed with the fractal geometry theory, from which the fractal dimension of the network is obtained. The result shows that the fractal dimension decreases with the threshold. Then, based on the R/S analysis method, it analyzes the time series of the clustering coefficient of network. The result illustrates that the time series show the long memory and persistence.Finally, the main contents and results of our research in this dissertation are summarized, as well as future research directions.