期刊名称：Quantitative and Qualitative Analysis in Social Sciences
印刷版ISSN：1752-8925
出版年度：2008
卷号：2
期号：1
页码：78-97
出版社：Quantitative and Qualitative Analysis in Social Sciences
摘要：EGARCH models for conditionally heteroscedastic time series have attracted a steadily increasingdegree of attention in financial econometrics and related fields. These models are able to representsome of the stylized features of financial returns, such us uncorrelation in levels but strong depen-dence in squares and log-squares and leverage e.ect. Estimation of these mo dels may be carried outby maximum likelihoo d and Whittle methods. This paper compares the finite sample behaviour ofboth methods, confirming that maximum likeliho od is more e.cient but bivariate Whittle sometimesperforms comparably
关键词：EGARCH; fractionally integrated EGARCH; maximum likelihoo d estimator; Whittle esti-mator; finite-sample properties
