期刊名称：Journal of Statistical and Econometric Methods
印刷版ISSN：2241-0384
电子版ISSN：2241-0376
出版年度：2020
卷号：9
期号：4
页码：53-86
语种：English
出版社：Scienpress Ltd
摘要：Recent studies have found indications oflong-range dependence in financial time series and used conventional,non-robust estimates of the memory parameter, which measures the degree oflong-range dependence, for the calculation of buy and sell signals. In thispaper, new robust estimators are proposed which are possibly more appropriatefor financial data. The new estimators are compared with various robust andnon-robust competitors by means of extensive simulations. In addition toadditive outliers and heavy-tailed distributions, also conditional heteroscedasticity is considered. The results show that the robust estimatorsdo not generally deliver better results than the conventional estimators butonly in special cases, the existing robust estimators with respect to theroot-mean-square error and the new robust estimators with respect to the bias.Finally, the different estimators are used to investigate possible long-rangedependence both in developed and developing stock markets. The results of this empirical study suggestthat long-range dependence is present only in the volatility and is thereforeof no use for directional forecasting and trading.
