Finite sample properties of a QML estimator of stochastic volatility models with long memory

被引:16
作者
Pérez, A
Ruiz, E [1 ]
机构
[1] Univ Carlos III Madrid, Dept Stat & Econ, Madrid, Spain
[2] Univ Valladolid, Dept Appl Econ Stat & Econometr, Valladolid, Spain
关键词
fractional integration; stochastic volatility; quasi-maximum likelihood estimator;
D O I
10.1016/S0165-1765(00)00373-6
中图分类号
F [经济];
学科分类号
02 ;
摘要
We analyse the finite sample properties of a QML estimator of LMSV models. We show up its poor performance for realistic parameter values. We discuss an identification problem when the volatility has a unit root. An empirical analysis illustrates our findings. (C) 2001 Elsevier Science BN. All rights reserved.
引用
收藏
页码:157 / 164
页数:8
相关论文
共 6 条
[1]   The detection and estimation of long memory in stochastic volatility [J].
Breidt, FJ ;
Crato, N ;
de Lima, P .
JOURNAL OF ECONOMETRICS, 1998, 83 (1-2) :325-348
[2]  
Ding Z., 1993, J. Empir. Finance, V1, P83, DOI [DOI 10.1016/0927-5398(93)90006-D, 10.1016/0927-5398(93)90006-D]
[3]   MULTIVARIATE STOCHASTIC VARIANCE MODELS [J].
HARVEY, A ;
RUIZ, E ;
SHEPHARD, N .
REVIEW OF ECONOMIC STUDIES, 1994, 61 (02) :247-264
[4]  
Harvey A.C., 1998, FORECASTING VOLATILI, P307
[5]  
HARVEY AC, 1993, EM93268 STICERD LOND
[6]   QUASI-MAXIMUM LIKELIHOOD ESTIMATION OF STOCHASTIC VOLATILITY MODELS [J].
RUIZ, E .
JOURNAL OF ECONOMETRICS, 1994, 63 (01) :289-306