Research On Network Entropy Of Stock Market And Its Influencing Factors

With the development of complex network theory,the use of complex networks to study financial markets has become an effective method.The use of network entropy method has become more mature in the study of financial markets.Shannon entropy,Renyi entropy,and Tsallis entropy have all been used to study the financial markets of various countries.However,the exploration of the factors of network volatility and related influence factors in China's financial markets is relatively lacking.An empirical study of the network volatility and influencing factors will provide a good guide to the design of real investment strategies and the avoidance of investment risks.Therefore,it is necessary to explore the network volatility of China's stock market and its influencing factors.Based on the data of China's Shanghai and Shenzhen Stock Markets,this dissertation first constructs a dynamic weighted network based on time series.It analyzes the average shortest path length,average clustering coefficient,average value and variance of side weights of the stock network,and finds the relationship among each index;The network entropy model analyzes the volatility of the stock market returns and studies its applicability.Based on this,it classifies industries based on new industry classifications.Finally,it analyzes the influence factors of network entropy to find out the impact of stocks.The indicators of the market rate of return fluctuations and their influence methods,through the identification and analysis of the factors affecting the network entropy,help deepen the understanding of network entropy,and then establish investment strategies based on relevant factors.After empirical analysis,the characteristics of the network structure of China's stock-associated networks are significantly linearly correlated with each other;network entropy can effectively characterize the volatility of the stock market,in which Tsallis and Renyi entropy can make ? take negative to characterize extreme situations;relatively speaking,Mining industry,real estate industry,and financial industry have better correlations between network entropy and return rate fluctuations in different stocks' operating cycles;fluctuations in average agglomeration coefficient,shortest path length,stock returns,and stock returns have a significant effect on the dynamic entropy of the network.Network entropy can describe the structure of the stock price fluctuations and the degree of stability of the model.The more stable the stock price volatility,the more effective the risk control can be in the investment.