Granger相关性与时间序列预测

被引:1
作者
王立柱 [1 ,2 ]
刘晓东 [1 ]
机构
[1] 大连理工大学控制科学与工程学院
[2] 沈阳师范大学数学与系统科学学院
关键词
时间序列; 预测; Granger相关; 神经网络;
D O I
10.13195/j.kzyjc.2013.0005
中图分类号
F224 [经济数学方法]; TP183 [人工神经网络与计算];
学科分类号
0701 ; 070104 ; 081104 ; 0812 ; 0835 ; 1405 ;
摘要
提出一种将Granger相关信息用于时间序列预测的方法,以解决时间序列预测过程中信息利用不完全的问题.首先,通过Granger相关性检验确定时间序列系统中的可利用信息;然后,利用神经网络将可利用信息抽取出来;最后,将抽取的可利用信息融入到时间序列的预测中.实验结果验证了所提出预测方法的有效性和稳定性.
引用
收藏
页码:764 / 768
页数:5
相关论文
共 9 条
[1]   Does CPI Granger-cause WPI? New extensions from frequency domain approach in Pakistan [J].
Shahbaz, Muhammad ;
Tiwari, Aviral Kumar ;
Tahir, Mohammad Iqbal .
ECONOMIC MODELLING, 2012, 29 (05) :1592-1597
[2]   Forecasting nonnegative option price distributions using Bayesian kernel methods [J].
Park, Hyejin ;
Lee, Jaewook .
EXPERT SYSTEMS WITH APPLICATIONS, 2012, 39 (18) :13243-13252
[3]   Adaptive neural network model for time-series forecasting [J].
Wong, W. K. ;
Xia, Min ;
Chu, W. C. .
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2010, 207 (02) :807-816
[4]  
Soft-computing techniques and ARMA model for time series prediction[J] . I. Rojas,O. Valenzuela,F. Rojas,A. Guillen,L.J. Herrera,H. Pomares,L. Marquez,M. Pasadas. Neurocomputing . 2007 (4)
[5]  
Unified structural equation modeling approach for the analysis of multisubject, multivariate functional MRI data[J] . JieunKim,WeiZhu,LindaChang,Peter M.Bentler,ThomasErnst. Hum. Brain Mapp. . 2007 (2)
[6]  
Interest rate linkages in the Eurocurrency market: Contemporaneous and out-of-sample Granger causality tests[J] . Zijun Wang,Jian Yang,Qi Li. Journal of International Money and Finance . 2006 (1)
[7]  
Mapping directed influence over the brain using Granger causality and fMRI[J] . Alard Roebroeck,Elia Formisano,Rainer Goebel. Neuroimage . 2004 (1)
[8]   FCM-based model selection algorithms for determining the number of clusters [J].
Sun, HJ ;
Wang, SR ;
Jiang, QS .
PATTERN RECOGNITION, 2004, 37 (10) :2027-2037
[9]   Modified Wald tests under nonregular conditions [J].
Lutkepohl, H ;
Burda, MM .
JOURNAL OF ECONOMETRICS, 1997, 78 (02) :315-332