Dynamic network topology and market performance: A case of the Chinese stock market

After the subprime mortgage crisis, plenty of abnormal market performance indicates that financial markets can be regarded as complex systems and it's time to break through some classical models. To tackle the issue, we propose novel complex networks methods to identify financial crises and explain some performance of the Chinese stock market. Firstly, we use the daily closing prices to construct the dynamical complex networks and their minimum spanning tree (MST) maps. Secondly, we characterize topological evolution of dynamical MSTs by employing normalized tree length, node degree distribution, centrality measures, node strength distribution and edge survival ratios. Furthermore, empirical analyses show that: (i) the normalized tree length can be used to identify financial crises, it declines sharply in the run‐up to, and during the financial crisis, and increases rapidly afterwards; (ii) the normalized tree length is positively correlated with market return and negatively correlated with market tail risk and volatility; (iii) the closeness centrality of most stocks is significantly negatively correlated with individual returns and positively correlated with individual volatility; (iv) the node degree and node strength in most of MSTs follow the power‐law distribution; (v) the edge survival ratio analysis indicates that the dependence structure of the Chinese stock market is relatively stable.