Research On Financial Market Asset Prices Co-movement And Risk Contagion

Currently, due to the frequent occurrence of financial crisis, the study on the risk contagion has become a hot topic. In the past, the multivariate statistical models occupied the most important position in analyzing the risk contagion. For example, a large amount of scholars employ the MGARCH models to test the volatility spillover effect to explore the crisis propagation. However, as we all known, for these multivariate statistical models, the "dimension disaster" is unavoidable. Thus, it is hard to use these models to accurately analyze the risk contagion among several financial markets. Fortunately, the birth of complex network science provides lots of convenience for our further investigation.From the view of complexity science by using the mature theory and methodology in complex network, I mainly focus two aspects. The first studying is to explore the impacts of financial crisis on the stock markets, and the other is to study the crisis contagion. The main works and innovations are given as follows:(1) We choose the daily closing price of 268 constituent stocks of the S&P500 index,221 stocks of London Stock Exchange,148 constituent stocks of the Shanghai Composite index and 152 constituent stocks of the Hang Seng index as the research objects to study the evolution of stock correlation in USA, UK, Chinese mainland and Hong-Kong respectively under the impact of the US sub-prime crisis. The obtained results are as follows:① For all the stock markets, during the crisis, the stock correlation strengthened significantly. This fact well interpret why the stock prices always grow or decline together during the system risk.②For USA, UK, Chinese mainland and Hong-Kong stock markets, the financial crisis changed the stock correlation structure, loosening the stock correlation firstly, and then tightening it again. From the perspective of complex network, the topology structure of the maximal spanning tree changes from compact star-like to loose chain-like first and then it turns to compact star-like once more.(2) Using a moving window to scan through every daily sector closing price over a period from Jan 2,2003 to Dec 26,2013 and transfer entropy to investigate the information flow among industry groups, we figure out the corresponding industry’s transfer entropy matrix in every given window. Next, a further analysis of information flow’s variations between industries is carried out. Finally, we also use the daily trading volumes of each industry group to have careful analyses on the trading volume correlation between sectors. The obtained results are as follows: ① The Financial industry sector is the arch-criminal of the US sub-prime crisis. ② Over the full-outbreak periods, the strength of information flow among industry groups reaches a peak.③ Before January 2007, the main information flow is from the Financial sector to Non-Daily consumption, Energy, Raw material and Industrial sectors, and the Financial sector is the dominant industry. However, around 12 April 2007, the main information flow has changed from Energy, Raw material and Industrial groups to Telecom, Daily consumption, Public utilities and Health care industries. Now, the most influential industries have turned to be Financial, Energy, Raw material and Industrial sectors.④The trading volume correlation is the fundamental reason of the crisis propagation.All the above studies could help us learn more about the financial crisis, meanwhile can be a good guide to the risk management of stock investment.