An Empirical Study Of Stock Market Data Analysis And Network Construction Based On Minimum Spanning Tree Method

With the development of economic globalization,the global financial activities are more and more closely linked,the uncertainty of financial market is increasing,showing a complex development trend.As the world's first developing country,China's financial market is in an important position,and the financial market is related to the stability of the entire financial market.The fluctuation of a key industry may lead to the fluctuation of stock price among multiple industries in the whole market,which may generate potential risks.With the development of the complex network theory,it is an effective way to study the stock market based on the volatility correlation between the stock prices of various industries,and the relevant research using the minimum spanning tree method is more and more mature.This paper uses this method to build a network of Chinese stock market and explore the internal evolution of the stock market.There are many literature reports about the structure and evolution of the world stock market or each stock index by using complex network,but few studies use partial association network to do the network topology of China stock market.Therefore,the research on the network topology of China stock market based on the minimum spanning tree method of association in this paper has certain innovation.Based on the complex network theory,this paper studies the Chinese stock market,and does the following work: 1.Select 262 stocks data of Shanghai and Shenzhen 300 index from June 2016 to June 2019,calculate the daily return series corresponding to the daily stock price of 262 stocks for three years,and analyze the data after presenting the stock price fluctuation trend chart.Pearson correlation coefficient is used to measure the stock relevance and analyze the relationship between stocks.A three-year undirected weighted network is constructed by using the minimum spanning tree method to dynamically evolve the network structure.It is found that the stock market network of Listed Companies in China has industry clustering.2.From the overall characteristics of network structure and the centrality of nodes,this paper analyzes the related network structure of Chinese listed companies,and finds that(1)in the whole network,the correlation between financial stocks is the strongest,and financial stocks are in the core position in the network.(2)Manufacturing industry,energy industry and high-tech industry play an important rolein the whole network and are closely related.Manufacturing industry and financial industry,energy industry and financial industry stocks have a strong correlation.3.Divide the network into communities,explore the evolution rule of stock market communities,and make statistics on the geographical distribution of nodes in different communities.It is found that there is regional linkage in stock volatility.If an abnormal stock occurs in a community with strong correlation,it may cause stock volatility in the region and affect the fluctuation of the whole network.The stock in the three urban agglomerations has a high impact on the market.The network structure of China's stock market is basically consistent with the key position of each stock in the actual situation of economic development.