Volatility Research And Dependence Analysis Of Financial Markets Based On CARR Model

With the development of modern information technology, financial theory and financial engineering technology, the circulation efficiency of international financial market is greatly improved. The varieties and speed of transactions are growing rapidly, sharply increased the volatility of financial market. The volatility research of financial market is becoming focus problem of academic research.Conditional auto-regressive range model(CARR) was proposed by Chou(2005),which characterizes the dynamic structure of range by combing range and GARCH model. He concluded that CARR model can produce more efficient volatility estimates as compared to the commonly adopted GARCH model. This dissertation will discuss the volatility of Shanghai stock market and the dependence of Sino-US market based on CARR and Copula function. The key points and main achievements of this work are listed as follows:1.Choose the CARR model which has the best goodness of fit through the empirical analysis, and then conducted an empirical study on data of Shanghai composite index based on selected CARR model and GARCH model.2.Threshold model and CARR model have been combined to construct nonlinear CARR model. Comparison of fitting effect between threshold CARR model and traditional CARR model have been done.3.Copula-CARR model is constructed and used study the degree and patterns of dependence between Shanghai stock market and American stock market.4.Select the 2008 global financial crisis as the cut-off point and compared the dependence in different stages between Shanghai stock market and American stock market.