Study Of The Gold Price Fluctuations In The Stochastic Volatility Model

In recent years China has been confronted with the huge inflationary pressures, due to the subprime mortgage crisis, the European debt crisis caused by the shock of the global financial markets and the domestic CPI and PPI high and severe shock. China’s individuals and institutional investors are faced with finding safe, stable investment vehicles. Experienced a certain level rise due to the uncertain economic situation worldwide and the depreciation of the dollar factors also led to the price of gold around the world. China has established ten years ago, the Gold Exchange and launched a variety of gold trading varieties, the gold futures market is also developing in recent years, these can be thought of our individual and institutional investors to participate in the secondary market of gold investment to provide the platform can also be achieved increasing the value of assets and risk aversion. However, in the gold market, investors will face the risk of gold price fluctuations, China’s gold market gold price fluctuation, it is necessary. This paper is based on the above context.The economical of the finance time series have the universal phenomenon of volatility, but the volatility is a core research question which to describe Gold market. At present, the model described in the financial markets, the volatility characteristics of two categories:one is autoregressive conditional heteroskedasticity (ARCH) model that Engle proposed in1982; the other is the stochastic volatility model that Taylor proposed in1986. In recent years the two models have developed very fast in our country, the researcher proposed lots of expanding model, for example EGARCH,GARCH-M, heavy-tail SV model, SV model in mean and so on. Theoretical circles on the volatility or heteroscedasticity found that the main features of the volatility can be attributed to the following aspects:①peak, fat tail;②volatility clustering;③asymmetry. Did China’s gold price volatility have these characteristics? This paper propose the significance of the research at first, then review the existing research and compare the usually used GARCH model and Stochastic Volatility model. Followed by analysis of gold price volatility factor theory, including the world gold market, gold demand and supply analysis, and the factors that affect the gold price fluctuations. The next selected the London gold market from October30,2002to August19,2011, the afternoon fixing price and the Shanghai the Au9995Gold spot price as a variable, Analysis of the statistical characteristics of these two variables volatility, this paper introduces the DIC criteria and the use of Bayesian methods, the pros and cons of the Comparison of GARCH and SV model in fitting the afternoon fixing price of London Bullion Market and Au9995gold spot price data, reached the following conclusions:The major conclusions of the part are listed below:(1)Asset returns in Shanghai and London gold market exhibits non-Gaussian distribution. There exhibits a well-known stylized fact in gold market--volatility clustering.(2)GARCH models are good descriptions of this time-varying volatility in gold returns. EGARCH(1,1)model can better described the fact than GARCH(1,1)model.(3)The model imply that volatility exhibits a high degree of persistence. The impact of new information placing on volatility in gold market will last a long time. So it indicates that it is necessary to develop gold future to control or manage the gold spot market risk caused by long term impact of new information.(4)The parameter for asymmetry in EGARCH model that recognizes the asymmetric behavior of volatility are highly significant positive in the case. The results demonstrate there is an asymmetry in the impact of news on volatility. Positive news surprises increase predictable volatility more than negative news surprises.(5)using the DIC criterion compared to the Shanghai and London gold price under the SV models system. The analysis discovered the SV-MT model was the best in analoging the London gold price,but the SV-T model was the best in analoging the Shanghai gold price.At last, the paper is conclusions and prospects. It is summarized the analysis of fluctuations in two models, and the special feature at the place of further study.