Research On Topological Characteristics Of Exchange Rate Network And Importance Of Node

In recent years, the complex network theory has been developed rapidly, which is widely used in sociology, biology, computer science, physics and other disciplines.The complexity of the economic system determines its nonlinearity, and the system is no longer a simple linear correlation between the factors,The traditional theory is more and more unable to explain the economic situation in reality, the introduction of complex network into the economic system is a great feat in the scientific community.Based on the complexity theory, this paper proposes a new perspective on the relationship between monetary and exchange rate fluctuations: the monetary system as a complex network system consisting of 47 interconnected monetary systems.The research of this thesis is divided into two parts:In the first part, the exchange rate network is constructed by threshold method and then we analyze the topological characteristics of the exchange rate.In this part, the characteristics of the degree and the degree distribution, clustering coefficient, the average path length of the exchange rate network are analyzed, and the market conditions of different periods are compared.Through the analysis of the average degree of the network, to reflect the changes of the exchange rate market fluctuations, when the average degree of network is high, shows that the volatility of the market volatility is relatively high, and we find that in the early and the last stage of the financial crisis,the Volatility correlation of the exchange rate market is relatively high.Through the analysis of the degree distribution, clustering coefficient and average path length,we find that the exchange rate is close to the small world network.On the basis of above,we analyze the correlation between the average degree of exchange rate network and the global economic index, we find that the economic growth rate and the global trade index have a negative correlation with the volatility correlation of the exchange rate market.Calculate the standard deviation of all monetary fluctuations in the network, and use the average value of the standard deviation of all monetary fluctuations in the same period to reflect the fluctuation of the exchange rate market, and compare to the average degree of the network, and find that the volatility correlation of the exchange rate market is lagging behind the volatility of the exchange rate market.In the second part, the importance of network nodes is evaluated and sorted, and the correlation between the average degree of the currency exchange rate and the exchange rate is tested.In this part, we analyze the node importance of the exchange ratenetwork which is constructed by the dynamic threshold, analyze the fluctuation of different currencies in the same time point.The importance of the nodes in a single network analysis can not show the volatility correlation of currency in the whole time series, so we have a statistical analysis of the importance of the nodes in the network in all quarters of the top six of the currency, and then a comprehensive analysis of its volatility.For the important nodes, the correlation between the economic variables and the average degree of exchange rate network is analyzed. The results show that the exchange rate of the most important nodes and the average degree of the exchange rate network have a negative correlation.