Abstract: In order to effectively avoid the fuel oil price risk in China,this paper develops a dynamic hedging model based on the minimum-variance principle; specifically,it uses the Copula function and its rank correlation coefficient to replace the linear correlation coefficient,and the dynamic standard deviations for both fuel oil spot and futures returns are obtained by means of the GED-GARCH models. Then the analyses for the hedging activities are conducted for China's fuel oil market during 2009-2012. The empirical results indicate that,compared with the traditional linear hedging model based on the minimum-variance principle,the dynamic hedging model here can significantly improve the effectiveness of fuel oil futures market; specifically,the hedging effectiveness of the dynamic model reaches 44.76% during the sample period,up 32.1 points compared with that of the traditional linear model.