Based on the heteroscedasticity of international gold price, the generalized auto-regressive conditional rnheteroscedasticity model for predicting international gold price is established, and the international oil price and the rndollar index are introduced as exogenous variables to compensate for the defects of the traditional time series model in rnthe external influence. By using the above model, this paper fits international gold average monthly price from Jan rn1986 to Nov 2009, and forecasts the date from Dec 2009 to Apr 2010. The results show that the method is better than rncompared models that don't consider heteroscedasticity and exogenous variables, in prediction accuracy. The model is rnhelpful to gold investors and producers.%基于国际黄金价格的异方差性,首先建立了国际黄金价格的广义自回归条件异方差预测模型,并在该模型中引入国际石油价格和美元指数作为外生变量,以弥补传统时间序列模型忽略外界影响因素的缺陷.采用该模型对1986年1月至2009年11月的月平均国际黄金价格进行拟合,并对2009年12月至2010年4月的月平均国际黄金价格进行了实证检验,结果表明,该模型的整体精度好于不考虑异方差性和未引入外生变量的模型,从而为黄金投资者和生产者的决策提供帮助.
展开▼