Abstract: The securities system is an important part of the financial system. It is of great theoretical and practical value to explore the relationship between the dynamic evolution of industry association of listed companies and the risk of securities market. This paper uses mutual information coefficient to establish the industry correlation network of China's listed companies, uses CAViaR model to measure the risk of China's securities market, and analyzes the network correlation, distribution structure and market wind combined with DCCA coefficient method and nonlinear Granger Causality test. Finally, the network characteristics of the two risk aggregation regions are compared and analyzed. The results show that:when the security market risk increases, the industry system of listed companies will be more related, and each industry will show a similar market trend. At this time, although the system structure tends to be closer to the “center-periphery” structure, the self-regulation function of the industry system of listed companies will weaken or even dissolve the impact of market risk on the industry association and the external morphological characteristics of the association structure have periodic shocks; when the correlation between industries is enhanced, it will accelerate the spread of risks to a certain extent. According to the above conclusions, this paper puts forward the relevant policy recommendations.

Key words: correlation, complex network, network structure, market risk, mutual information coefficient