Research On Long Memory In Chinese Copper Future Market

As a high development of commodity economy, future transaction plays a significant role in our capital market, which is because of its special economical function. But its drastic changes also influence our economy deeply. In order to keep a healthy economical environment, it's the time to study the price behavior of future.As a main feature of price behavior, long memory breaks through the hypotheses in Efficient Market Hypothesis and gives a new study direction for capital pricing and risk management. Therefore by some nonlinear research methods, this paper analyzes empirically the long memory in the returns and volatilities of copper futures, which are for well understanding of the volatility behavior in Chinese copper futures market.Firstly, this paper explains some relevant theories and research statements, and the emphasis is put on the introduction to a series of testing methods and estimation means of long memory process. Secondly, empirical analysis with three analysis methods is carried out on the returns and volatilities of the whole copper future series.Third, on the basis of ICSS method, the long memory feature of every phase series is tested. At last, starting from the volatility modeling of long memory process, comparisons are made between GARCH model and FIGARCH model。Through this research, the long memory feature is demonstrated in our copper futures market. Getting rid of the switching regimes factors, the long memory feature of every phase series is reduced significantly. The FIGARCH model gives a better description and forecast of volatility in Chinese copper futures market.