A Bayesian MCMC based estimation of Long memory in state space model

被引:0
作者
Li, Yushu [1 ,2 ]
机构
[1] Norwegian Sch Econ, Dept Business & Management Sci, Bergen, Norway
[2] Lund Univ, S-22100 Lund, Sweden
来源
INTERNATIONAL WORK-CONFERENCE ON TIME SERIES (ITISE 2014) | 2014年
关键词
fractional difference; state space model; Kalman filter; Metropolis-Hastings algorithm; LOCAL WHITTLE ESTIMATION; TIME-SERIES; REGRESSION;
D O I
暂无
中图分类号
F [经济];
学科分类号
02 ;
摘要
To estimate the long memory series in the framework of state space model is rarely documented although the theoretical foundation was well built in late 90s, and the literatures concentrate mainly on the estimation in stationary case. This paper aims to estimate the parameters in a wide range of long memory series by applying approximate Maximum Likelihood Estimation (MLE) and Bayesian Monte Carlo Markov Chain (MCMC) methodology. We show that both methods perform quite well with the exception in the case that the series is nearly non-stationary, where pure MLE gives out seriously over biased estimation and the Bayesian MCMC estimation can avoid this problem when pre-knowledge is available.
引用
收藏
页码:1341 / 1352
页数:12
相关论文
共 28 条
[1]  
[Anonymous], NEW PALGRAVE DICT EC
[2]   FITTING LONG-MEMORY MODELS BY GENERALIZED LINEAR-REGRESSION [J].
BERAN, J .
BIOMETRIKA, 1993, 80 (04) :817-822
[3]  
Beran J, 1994, Statistics for Long-Memory Processes
[4]  
Besag J, 2004, IMA VOL MATH APPL, V138, P247
[5]  
Brockwell P. J., 1991, Time Series: Theory and Methods Springer Series in Statistics, V2nd
[6]  
Brooks SP, 1998, J ROY STAT SOC D-STA, V47, P69, DOI 10.1111/1467-9884.00117
[7]  
Chan NH, 1998, ANN STAT, V26, P719
[8]  
Durbin James, 2012, Time Series Analysis by State Space Methods
[9]   LARGE-SAMPLE PROPERTIES OF PARAMETER ESTIMATES FOR STRONGLY DEPENDENT STATIONARY GAUSSIAN TIME-SERIES [J].
FOX, R ;
TAQQU, MS .
ANNALS OF STATISTICS, 1986, 14 (02) :517-532
[10]  
Geweke J., 1983, Journal of Time Series Analysis, V4, P221, DOI [10.1111/j.1467-9892.1983.tb00371.x, DOI 10.1111/J.1467-9892.1983.TB00371.X]