Parameter Estimation and Its Application of H on Control Factor with R/S Analysis
Wei Long Sun · Tongji yu xinxi luntan · 2010
R/S analysis is the main methods to reveal the long memory of financial time series.In order to improve the classic R/S and modified R/S analyses its deficiencies,this paper designed a R/S statistic with control factors,and applies Monte Carlo simulation to show that the improved method is more effective than the classical R/S with respect to the estimation of H index.The new method is applied to analyze the long memory of the return of Shanghai Composite Index and Shenzhen Component Index and average length of non-cyclical of these two indices.The results show that both Shanghai and Shenzhen stock market's return series have long memory,but for the Shanghai stock market's return series,there is no obvious average length of non-cyclical,while for the Shenzhen stock market's return series,there exists a average length of non-cyclical,which is about 308 days.