Research On Interactive Impact Of CSI Industry Index Return Rate Based On Dynamic Influence Model

In the context of the current rapid development of technology,the field of machine learning has developed rapidly.The use of machine learning to process and analyze economic data is not only a current research hot issue,but also can be used for investment behavior.With the rapid development of China’s securities industry,China’s securities market has become a very important field of national investment,and more and more scholars have conducted research around this investment market.Therefore,the main content of this study is to select 10 CSI industry index data from January 5,2015 to December 29,2019,based on the idea of Bayesian network,adopting an improved from the coupled Markov model The dynamic influence model,combined with the network connection diagram,explored the mutual influence relationship between the index returns of bull markets and bear markets at different periods in 10 industries.The results of the study show that from 2015 to 2019,the model found that the degree of mutual influence between the markets is relatively high.The dynamic change of the index can be used to roughly judge the possible change in the index return.We found that the strength of the interaction between the index returns of 10 industries in different time periods is different.In the rising market in 2017 and 2019,the consumer industries have shown a relatively high degree of influence on other indexes,while the utilities and energy industries are the influence receivers.In the decline in 2018,the raw materials and pharmaceutical industries were the issuers of market influence,while the financial and energy industries were the main influence receivers.Based on this,we find that investors should choose the issuer in the rising market,and the receiver in the falling market.