Financial contagion behavior analysis based on complex network approach

Recently, financial crisis occurred frequently, such as Asian crisis, sub-prime loan crisis and so on. And, financial contagion behavior analysis is an important subject in the field of international finance. This paper analyzes the financial contagion behavior of US subprime crisis and describes the co-movement behavior between US stock index return and the stock index returns of tested countries by complex network approach. The co-movement modes are defined by a coarse graining procedure, and then by weighted complex network models and evolutionary models transformation characteristics between the modes are analyzed. In order to analyze the sub-prime loan crisis contagion from US to other countries, we selected three main stock market indexes (China, Japan and Germany), S&P 500 index to present an empirical analysis, and the stock market index data is from 2006.1.1 to 2009.3.31. To analyze the contagion effect of the sub-prime crisis, the data is divided into three segments to be analyzed based on complex network approach. The empirical result shows that the co-movement modes of the S&P 500 index and the stock indexes of other countries are clustered around a few critical modes during the evolution. The co-movement modes have the characteristic of grouping, and the conversion of the co-movement modes requires an average of 3–5 days. This paper confirms the sub-prime crisis contagion behavior by comparing the parameters including degree, clustering coefficient and betweenness centrality in three stages. This research may provide further information between S&P 500 index and other stock indexes for the co-movement. More importantly, a new approach is provided for testing the financial contagion effect in this paper.