Research On The Statistical Measure Of Co-movement Among The World's Stock Markets

In the background of global economic integration,the economic exchange between main countries of the world is becoming more and more close.As a macroeconomic barometer,the co-movement of stock market has become the focus of many researchers.Studying on the stock market has important influence and practical significance.On the one hand,it can provide evidence for decisions of the government and ensure the healthy operation of the stock market.On the other hand,it can improve the return of investment and reduce the investment risk.However,the existing researches ignored the thought of measure method and did not discuss the applicability of the method that they had chosen.This phenomenon is easy to result in an error that different measure methods can get different results when researching the same object.Therefore,focusing on the theoretical and empirical research on the popular measure method of Co-movement,this article wants to find a more appropriate method in the specific condition,so that the empirical results have a reference value.The train of thought of this paper is as follows:First of all,researching on the existing literature,this article finds several measure methods of co-movement that have been used most.The methods include correlative coefficient analytical method,cointegration test,VAR model,Granger causal test and ARCH model.Secondly,this paper comparatively analyses the precondition,theoretical derivation process,scope of application and focuses on the analysis of the differences among the various measure methods.Thirdly,based on the daily closing data from January 1,2013 to May 20,2016,this article empirically analyses the co-movement of stock market of Chinese mainland,Japan,the United States and Hong Kong from four different dimensions.Through the research on the measure of co-movement,this paper obtains the following conclusions.Firstly,different measure methods are suitable for different purposes.Among them,the cointegration test is suitable for measuring the long-term equilibrium effect of stock market,VAR model is suitable for measuring the short-term impact of stock market,Granger causality test is suitable for measuring the causality effect of stock market,ARCH model is suitable for measuring the volatility transmission effect of stock market.Secondly,there is a degree of differences among the results of the same study about the measure of co-movement from different dimensions.Among them,Shanghai Composite Index and the Nikkei index have long-term equilibrium effect;there is a long-term equilibrium effect between Shanghai Composite Index and the Dow Jones industrial average,and the latter has a one-way short-term impact and volatility transmission effect on the former,the latter also has Granger cause for the former;Shanghai Composite Index and Hang Seng Index has a degree of Granger causality in both directions,and the former has a short-term impact on the latter,the latter has volatility transmission effect on the former.