| The volatility, as a quantitative uncertainties description of time series, is a crucial factor which can affect the financial derivatives pricing models, portfolio risk management and hedge strategies and other financial investment activities. To build a volatility model which can simulate the disciplinarian of financial market effectively, will strengthen the financial markets risk management, prevent the manager and the investor from financial crisis, maintain a healthy and stable development of the market, and promote economic prosperity. It has a very important theoretical significance and practical value to do some research on volatility.China’s stock market, which is an important part of the international financial market, has developed very rapidly in recent years, but many of the problems associated with its development is still in the exploration of various aspects of the rules and regulations are also not perfect, vulnerable to external factors deviate from the normal volatility, thus showing a high degree of instability. In addition, China’s stock market investors in the stock market’s performance are not rational, who can be easily affected by the surroundings, thus resulting in increased stock market asymmetry. Imperfect and investors are not rational system is that characterized by the prominence of China’s stock market, this feature makes the stock market is far greater than the risk of foreign mature stock markets. Therefore, there is an urgent need to study China’s stock market volatility, fluctuations sum up its rules for financial markets to improve market management regulations as a reference, but also provides some decision-making basis for investors to invest in stocks, thus contributing to China’s stock market which can be healthy stable development of healthy and stable development of the national economy.However, with the current international economic integration process, the economic ties between countries increasingly close linkage between the stock market and also have a contagious effect on all aspects of the stock market resulting in increased response, stock market returns in the stock market of a message time occurrence jump in or drop, resulting in volatility structure mutations. If the stock market structural breaks, it will have a significant impact on the stock market linkage and infection. If you do not consider the structure of the mutant stock market volatility at the time of the study, it would be inaccurate to describe the stock market volatility; if we consider the structure of the mutation, but the structure of point mutations or inaccurate determination of the high volatility state mistaken for low volatility state, managers and investors will make decisions ineffective or even significant losses, thereby reducing the risk management level. Therefore, the structure of volatility study mutations in the moment it is very important and urgent, but also on the volatility model chosen will require higher and higher.So, if you want to regulatory authorities and investor services to help managers improve the risk management of China’s stock market, to ensure the healthy and stable development of the stock market; and if you also want to assist investors with the ability to cope with risks and ensure investment interests, reduce losses, volatility in the pair, not only do we need to study when considering the structure of the mutant characteristics, but also on the structure of the mutant characteristics accurately portrayed.In view of this structural break in the stock market, many scholars in various ways to capture a variety of models, but it is not a huge breakthrough until Hamilton and Susmel(1994) Application of the Markov transition mechanisms in volatility. Because Markov regime switching(Markov Regime Switching, MRS) model can describe the different stages, status, economic behavior has different characteristics under the mechanism, you can just make up the traditional model which cannot reflect the GARCH mutation exhibit more volatility due to structural deficiencies market status and there are easily combined with other features of the model, it will be combined with the volatility model to describe the structure of the mutant volatility characteristics. However, Markov switching model when the model parameter estimation, the need for subjective volatility state divided, so that the final model parameter estimation results cause some interference, which makes the model on the market fit ineffective.HMM model is widely used in pattern recognition, this model consists of two Markov process sequence from the double random process, and be able to describe the deep which can cause the observed sequence appears related information, so often used to detect some of the causes of an Hidden events occurring state transition probabilities and other relevant information. Based on this feature, combined with the performance of the volatility that we can see but cannot carry out structural mutation status volatility well divided, we introduced the HMM volatility model, build a new model of the structure of China’s stock market mutation characteristics capture and portray.For the sample of choice, we selected the most representative of the overall trend of China’s A-shares in Shanghai and Shenzhen 300 Index. CSI 300 Index, which contains a large scale and covers about 60% of the Shanghai and Shenzhen stock market, has a good representative of the market. And it uses the most reliable index of the current scheduling technology to ensure the reliability and continuity of data. So we choose the CSI 300 Index as the sample data of stock prices fluctuations in this paper,In summary, this paper will study the volatility of the most classic GARCH models, combined with the hidden Markov model HMM-GARCH volatility model to expand its study in detail the system. In order to find out the best fit for the market model, we also estimate the parameter of GARCH model and MRS-GARCH model.The empirical result shows that the average return of the stock market was positive, and the volatility of the stock market there is a very significant positive correlation with income; GARCH model can characterize the aggregation of volatility and the fat tail, but cannot simulate the stock Market mutation; HMM-GARCH model combined with conditional heteroscedasticity can portray the volatility characteristics of the three at the same time; meanwhile, it also can capture the stylized facts volatility structural breaks, and the likelihood value display HMM-GARCH model fitting degree sequence of the highest yields; comparative study with the actual rate of return can be seen, HMM-GARCH model fitting out of the state diagram basically the actual trend of the market fluctuations think fit, and can basically find structural break point of the market.Subsequently, based on the optimal model, combined with the global economic situation and changes in our policy recently, from a macroeconomic point of view to analyze the volatility of stock prices, it can be concluded: Investment in the stock market in general is to get income; most of the benefits gathers in the part of short-term speculative behavior; China’s stock market ties closely with the international financial markets, by the combined effects of the global various factors. |
PDF Full Text Request