收稿日期: 2016-03-16
网络出版日期: 2017-10-30
基金资助
国家自然科学基金资助项目(11371242)
Bayesian estimation of autoregressive models with time-varying coefficients
Received date: 2016-03-16
Online published: 2017-10-30
陈云仙, 高星月, 王钰莹, 何幼桦 . 时变系数自回归模型参数的贝叶斯估计[J]. 上海大学学报(自然科学版), 2017 , 23(5) : 732 -741 . DOI: 10.12066/j.issn.1007-2861.1754
This paper analyzes a time-varying autoregression model where coeffcients are correlated at different time. When only one sample path is chosen, the Bayesian method is used for estimation. Formulas of estimation of the first order model are presented. This paper also discusses how the estimation is affected by the coeffcient values and the length of samples. To conclude, based on an empirical evidence, it is shown that the statistical results are consistent with the actual data.
[1] Dahlhuaus R. Fitting time series models to nonstationary processes [J]. The Annals of Statistics, 1997, 25(1): 1-37.
[2] West M, Prado R, Krystal A D. Evaluation and comparision of EEG traces: latent structure in nonstationary time series [J]. Journal of the American Statistical Association, 1999, 94(446): 375-387.
[3] Prado R, Huerta G. Time-varying autoregressions with model order uncertainty [J]. Journal of Time, 2002, 23(5): 599-618.
[4] Bibi A, Francq C. Consistent and asymptotically normal estimators for cyclically timedependent linear models [J]. Annals of the Institute of Statistical Mathematics, 2003, 55(1):
41-68.
[5] Francq C, Gautier A. Estimation of time-varying ARMA models with Markovian changes in regime [J]. Statistics and Probability Letters, 2004, 70(4): 243-251.
[6] Triantafyllopoulos K, Nason G P. A Bayesian analysis of moving average processes with time-varying parameters [J]. Computational Statistics and Data Analysis, 2007, 52(2): 1025-
1046.
[7] Nakajima J, Kasuya M, Watanabe T. Bayesian analysis of time-varying parameter vector autoregressive model for the Japanese Economy and Monetary Policy [J]. Journal of the Japanese and International Economies, 2011, 25(3): 225-245.
[8] Ijima H, Grivel E. Deterministic regression methods for unbiased estimation of time-varying autoregressive parameters from noisy observations [J]. Signal Processing, 2012, 92(4): 857-871.
[9] 王永民, 何幼桦, 忻莉莉, 等. 时变自回归模型系数的估计及预测[J]. 应用数学与计算数学学报, 2007, 21(2): 35-41.
[10] Bekiros S. Forecasting with a state space time-varying parameter VAR model: evidence from the Euro area [J]. Economic Modelling, 2014, 38(C): 619-626.
[11] Barnett A, Mumtaz H, Theodoridis K. Forecasting UK GDP growth and inflation under structural change: a comparison of models with time-varying parameters [J]. Bank of England
Quarterly Bulletin, 2012, 89(3): 1237-1246.
[12] 中华人民共和国国家统计局National data—–国家数据[EB/OL]. [2015-12-30]. http://data.stats.gov.cn/easyquery.htm?cn=C01.
/
| 〈 |
|
〉 |
