Empirical Analysis On Long Memory Character Of Chinese Stock Market Volatility

As the development of finance theory,modern risk management,modern investment and portfolio theory, and deriving products and tools, stock market volatility become more and more important. This paper mainly studies the long memory character of the stock market volatility, and applies this character to compute Value at Risk of stock market index.This paper consists of four parts: general discussion (chapter 1~2); long memory analysis of stock market volatility in China (chapter 3~5); using high frequent data studies stock market volatility and its applications (chapter 6); finally, summary of whole paper is given. The detailed content is as follow:Chapter 1 explains the background and survey of the long memory research; and chapter 2 introduces in detail the long memory character's definition, typical model, method of test and estimation, and the main cause of this character.Chapter 3 discuss the selections of the volatility's index series firstly, and then tests and estimates this long memory character of volatility in China; Chapter 4 introduces regime change identification technology-ICSS method, and then studies the relations between the long memory character and regime changes, in order to explain the cause of this character; Chapter 5 applies several models whose abilities to capture the memory of volatility is deferent to model, forecast and compute the VaR. Chapter 6 introduces the method of realized volatility constructed by high frequent data, then uses this method to study the characters of stock market volatility, and selects the most suitable model for realized volatility, finally, applies this model to forecast the VaR of stock market index.Chapter 7 sums up the whole paper, and discuses the following researches in the future.