Research On International Securities Market Relevance Affected By Sub-prime Crisis

The U.S. subprime mortgage crisis which triggered in 2007 as an information shock, gives a major impact on the global economy. Relative to the real economy, the performance of the stock marcket was faster and more efficient. So this paper mainly discuss the relevance between stock market around the world with the U.S. market.This paper use Granger causality test based on VAR model and the Copula function, these two models in different countries to analyze the impact of U.S. subprime mortgage crisis to the security market countries in the world. In this paper, we devide all the world stock market fall into two categories, one for the developed markets, another for the emerging markets;at the same time we also take time period into account discussing different times with different relevance. We choose the United Kingdom, France, Germany, Canada, Japan, Singapore, Hong Kong stock market as developed market; China, India, Russia and India as emerging market. In the analysis of different types of securities association with the U.S. market, this paper also considering the different relevance between different regions, such as the European and Asian markets with the U.S. market.The second chapter gives the literature review of defination on relevance between market , crisis transformation and the two models. Chapter III give the theoretical foundation, concepts, properties, formulas and applications of the two models. Chapters IV and V of this article with the structured method, the use of VAR and Copula functions two ways to make empirical test on market association. Each chapter briefly from the model, the empirical results to the analysis, completely give the whole process and description. Chapter VI discuss the transmission mechanism in theory and the compare the results based on actual results. Last part includes the overall conclusion of this chapter and the direction of future research.This article has three innovations in all.One innovation of this article is to classify the different types of securities markets respectively compared the market characteristics and geographical features from both the more developed markets and emerging markets, and the different characteristics between Europe and Asia marcket. The second innovation is to broaden the period of crisis as the pre-crisis then and after-crisis three periods. The third innovation is that were based on both the Granger causality test and VAR Copula function;although the principles and focus of the two methods are different, but similar conclusions can be given.So that we can compare the result with reality give more accurate conclusion.Data period was from January 2005 to December 2009 and we take the 12 market index returns as samples. After the comparison with U.S. stock price index we can get the relevance of different types of markets.The correlation coefficient can be obtained by a linear relationship between variables, VAR method can be obtained through the Granger causality analysis of the causal relationship between variables. Copula function sucked integration of different variables to create a function to obtain the correlation between different variables, which is not limited to linear. Meanwhile, the Granger causality test based on VAR impulse response function focus on the dynamic adjustment process. The Copula function may be only focus on static relevance. Therefore, we can roughly think of the Granger causality test based on the VAR is more emphasis on the market's " Spread " and Copula function is more focused on the market's "relevance."Granger causality test is based on VAR model. The VAR model estimation use the least square method which demands data stationary, therefore the unit root test for the data is needed.,So after that we can make Granger causality test and impulse response function analysis. When metioned Copula function,we first need to describe and characterize the distribution of the residual sequence of returns,which mainly through the GARCH (1,1) model to complete. Marginal distribution of the residual terms can represent the characteristics of return series, so this step also yields the grasp of their specific sequence. Copula function for the variables required to be in (0.1), with which the second step need to get the residual sequence of probability calculus. After The second step the Copula function can be used to calculated the correlation. Of course, how to choice Copula function also depends on the what kind of data you want to describe and the characteristics or background of the Copula function .Different types of Copula function characterize different markets. In this paper, samples is divided into three periods, a period of a bull market, two periods of bear market, so we choose Gumbel Copula function for the first step and Clayton Copula function for the others.Summarize the two models, we can gives the conclusions as follows:First, the U.S. subprime crisis brought about large fluctuations on the both U.S. securities market and global marckets.Second, different markets, the market's fluctuations transmission from U.S. to developed markets is stronger than the three emerging markets -- China, Brazil, India.Finally, among the same types of markets, the transmission efficiency of different regions are different:the impact of European market --- the United Kingdom, France, Germany,are faster than the Asian stock market such as Japan, Hong Kong, Singapore. As well as although the emerging markets, Russia's transfer is more visible. In short, the relevance of the stock market between United States and Europe is stronger than between the U.S. and Asian markets,and greater relevance between developed markets than emerging markets. Finally, we use classical theory such as arbitrage pricing model, efficient market theory , information asymmetry theory as a starting point for thinking, by contacting with actual situation to explain the rationality and effectiveness of the result.