Abstract: In this paper, relatively active contracts in China's futures market are selected from January 2002 to June 2011 as the samples, to test their day-of-the-week effect of the rate of return, conditional volatility, trading volume and open interest. In order to test the existence of the day-of-the-week effect and mode of existence more comprehensive, three kinds of distribution assumptions are introduced to portray the disturbance term, such as normal distribution, Student-t distribution and generalized error distribution(GED), to modify the traditional GARCH model. The empirical results show that the day-of-the-week effect of China's futures market is affected by its investor structure, especially individual investors, and its test results is associated with the research methods, in which the error term distribution directly affects the results. It also shows different futures in different sample periods has different day-of-the-week effect modes. But in the early stages of rectification in the futures market, the day-of-the-week effect has no found, and subsequently showed the positive Monday effect or reversal of the Monday effect. If the rate of return is different, the day-of-the-week effect mode is not the same.

Key words: China's futures market, day-of-the-week effect mode, error term distribution