The empirical analysis of long-term memory of stock market returns of China

Analysis of the effects of long-memory in security market returns has important sense in research of market efficiency and conforming framework of nonlinear system. Based on fractal market hypothesis, some hypothesis, some typical characteristics such as the long-term memory structure and non-periodic cycle of volatility and non-Gaussian distribution of return are discovered by R/S analysis in China stock markets. It is proved that the fractal market hypothesis can interpret better the nonlinearity inhere in China stock markets than efficient market hypothesis.