| With the gradual and further of financial globalization. Since the 1980s,The development which like financial liberalization, information technology, and financial innovation is the development of the globalization of financial markets is to promote the objective power. After the collapse of Brighton Forest System, the Federal Reserve in the Green Lionel period of low interest rate policy as well as related financial and opening up the signing of the agreement makes the degree of openness of financial markets continues to deepen. A large number of floating capital quick and free flow in the world. The continued flow of capital to promote financial deepening, financial innovations to accelerate development and enhance the efficiency of financial markets. However, at the same time the continued flow of capital also brought the global financial volatility and even the scope of the current financial crisis. As a result of economic globalization, the interdependence of the global economy also enhance, global financial market fluctuations in the price of linkage to any region of the local fluctuations in the financial markets will quickly spread, infection, enlarged to other regions of the world. In fact, along with the deepening of financial globalization, the world has taken place in several major financial crises, such as the 90's financial crisis in Southeast Asia, the current sub-loan crisis by the United States caused by the global economic crisis. Since the early 90's , China stock market up and down, and the regulation of financial markets has been constantly strengthened and improved after the first time crisis. People in this financial crisis not only loss wealth but also for all on a financial risk management on the importance of the special lectures.In this paper, We used econometrics methods empirical analysis the correlation of two of China's major stock markets as well as between the mainland and Hong Kong stock market. Traditional relationship researches of different financial markets are the use type of CC-GARCH methods. In this paper, we the use of Copula methods, the introduction of Song et al (2005) proposed methods of MBP, as well as Fan (2007) proposed methods to estimate and test MMBP analysis the correlation of different financial markets. The broad structure of the article as follows: First of all, given the necessary theoretical explanations ,and then introduced the Copula theory, the next introduced by Song et al (2005) proposed methods of MBP and Fan (2007) proposed methods MMBP. By Monte Carlo simulation, based on the conclusion that the MBP methods and MMBP method Copula function more accurate than the IFM estimated that is traditional method. Through empirical analysis correlation which China's two stock markets and between the mainland and the Hong Kong stock market and put forward economic policy implications are given by the empirical analysis.In the first chapter, we outlined the financial econometrics and financial time series model of the development. By recalling the development of financial econometrics and financial globalization leads to financial risk management, especially among financial time series analysis of correlation, and then introduced univariable financial time series analysis model of the main categories such as ARCH models, SV models.In the second Chapter,we introduced the Copula method. The greatest advantage of Copula function is the edge of the distribution of different time-series connected, and analysis of the correlation between the two. This is a great expansion of the multivariate time series research, especially in multivariate time series on the relevance of research. In many cases, we often need to analyze different economic market, the stock market, futures market and other related time series similar to the relevant characteristics analysis to obtain the portfolio of financial assets and so reduce the financial risk research. First described in Copula function definition and the basic nature. Copula function is one of the most basic nature can be different from the marginal distribution function of the time series to connect through the Copula Function and consistency will be the correlation coefficient. Copula function under different circumstances can be divided into different categories of Copula functions, the main group can be divided into elliptical Copula function, Copula Archimedean family function, as well as the extreme Copula function. The most widely used is the elliptic function Copula family Gaussian Copula functions and student-t Copula function. In this article we have introduced more or less done and the corresponding two-dimensional or three- dimensional scatter plot.In the third chapter, we introduced methods of MBP by Song et al (2005) proposed, as well as Fan (2007) proposed MMBP method. Mainly, MBP method applied to the low correlation between the time series model analysis, But,when highly relevant in order to solve the time-series estimation problem, Fan and others to improve the methods of MBP proposed method MMBP. In this paper, these two methods with the traditional estimation methods Copula compared IFM theoretical analysis and methods of derivation introduced. And through Monte Carlo simulation methodology MBP and MMBP method is superior to the traditional conclusion of the IFM.In the fourth chapter, we empirical research the correlations of china's two major stock markets as well as China's mainland and Hong Kong stock market. In this paper, based on the actual situation of China's stock market, China's stock market return has a fat tail propriety does not meet the normal distribution, So ,we used the MMBP method and the marginal distribution of student-t distribution on an analysis of China's stock market, and that the method is more applicable to China's national conditions.Finally, the paper summarizes empirical results. We give the explanation of the relationship of China's stock market, the correlation between such a reasonable explanation, and finally proposed the corresponding policy recommendations. |
PDF Full Text Request