[1]杨淳沨,吴国成,伍家松,等.基于gAIC和滑动窗的自回归模型参数估计算法[J].东南大学学报(自然科学版),2018,48(3):381-386.[doi:10.3969/j.issn.1001-0505.2018.03.001]
 Yang Chunfeng,Wu Guocheng,Wu Jiasong,et al.Parameter estimation algorithm for autoregressive model based on gAIC and moving window[J].Journal of Southeast University (Natural Science Edition),2018,48(3):381-386.[doi:10.3969/j.issn.1001-0505.2018.03.001]
点击复制

基于gAIC和滑动窗的自回归模型参数估计算法()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
48
期数:
2018年第3期
页码:
381-386
栏目:
计算机科学与工程
出版日期:
2018-05-20

文章信息/Info

Title:
Parameter estimation algorithm for autoregressive model based on gAIC and moving window
作者:
杨淳沨吴国成伍家松姜龙玉孔佑勇舒华忠
东南大学计算机网络和信息集成教育部重点实验室, 南京 210096; 东南大学计算机科学与工程学院, 南京 210096; 东南大学中法生物医学信息研究中心, 南京 210096
Author(s):
Yang Chunfeng Wu Guocheng Wu Jiasong Jiang LongyuKong Youyong Shu Huazhong
Key Laboratory of Computer Network and Information Integration of Ministry of Education, Southeast University, Nanjing 210096, China
School of Computer Science and Engineering, Southeast University, Nanjing 210096, China
Centre de Recherche en Information Biomédicale sino-français, Southeast University, Nanjing 210096, China
关键词:
滑动窗 gAIC 最佳参数搜索 自回归模型 参数估计
Keywords:
moving window generalized Akaike information criterion(gAIC) optimal parameter search autoregressive model parameter estimation
分类号:
TP399
DOI:
10.3969/j.issn.1001-0505.2018.03.001
摘要:
为了提高自回归模型参数估计的准确性,提出了一种基于通用赤池信息准则和滑动窗的自回归模型参数估计算法.首先,使用通用赤池信息准则估计自回归模型的模型阶数,初步得到自回归模型中选项的候选向量集;然后,针对该候选向量集中的各候选项,采用滑动窗法获得其相应的权重值;最后,根据各候选项的权重值,利用自适应的最佳参数搜索算法进一步剔除候选向量集中的干扰候选项,得到自回归模型的最终模型选项及相应的模型系数值.实验结果表明,对于不同长度的信号,不同组合实验方案下所提算法获得的正确率最高,接近于90%.
Abstract:
To improve the parameter estimation accuracy of autoregressive(AR)models, a parameter estimation algorithm for AR model based on the generalized Akaike information criterion(gAIC)and the moving window is proposed. First, the gAIC is used to estimate the model order of the AR model, and the candidate vectors of the item of the AR model are obtained preliminarily. Then, the weighted value of each term of the candidate vectors is obtained by using the moving window. Finally, according to the weighted values of the candidate terms, the interferential items of the aforementioned candidate vectors are eliminated by the adaptive optimal parameter search algorithm. The final model items and the corresponding model coefficient are obtained. The experimental results for different combined experimental plans show that for different signal lengths, the accuracy of the proposed algorithm is the highest and almost reaches 90%.

参考文献/References:

[1] 蒲慕明, 徐波, 谭铁牛. 脑科学与类脑研究概述[J]. 中国科学院院刊, 2016, 31(7): 714,725-736. DOI: 10.16418/j.issn.1000-3045.2016.07.001.
[2] Sporns O. Brain connectivity [EB/OL].(2007-10-27)[2017-05-10]. http://scholarpedia.org/article/Brain_connectivity.
[3] Granger C W J. Investigating causal relations by econometric models and cross-spectral methods [J]. Econometrica, 1969, 37(3): 424-438. DOI: 10.2307/1912791.
[4] Kamiński M J, Blinowska K J. A new method of the description of the information flow in the brain structures[J]. Biological Cybernetics, 1991, 65(3): 203-210. DOI: 10.1007/BF00198091.
[5] Baccalá L A, Sameshima K. Partial directed coherence: A new concept in neural structure determination[J]. Biological Cybernetics, 2001, 84(6): 463-474. DOI: 10.1007/PL00007990.
[6] Friston K J. Functional and effective connectivity:A review[J]. Brain Connectivity, 2011, 1(1): 13–36. DOI: 10.1089/brain.2011.0008.
[7] Akaike H. Power spectrum estimation through autoregressive model fitting[J]. Annals of the Institute of Statistical Mathematics, 1969, 21(1): 407-419. DOI: 10.1007/BF02532269.
[8] Rissanen J. A universal prior for integers and estimation by minimum description length[J]. The Annals of statistics, 1983, 11(2): 416-431. DOI:10.1214/aos/1176346150.
[9] Khorshidi S, Karimi M, Nematollahi A R. New autoregressive(AR)order selection criteria based on the prediction error estimation[J]. Signal Processing, 2011, 91(10): 2359-2370. DOI:10.1016/j.sigpro.2011.04.021.
[10] Zhao Y, Billings S A, Wei H L, et al. A parametric method to measure time-varying linear and nonlinear causality with applications to EEG data[J]. IEEE Transactions on Biomedical Engineering, 2013, 60(11): 3141-3148. DOI: 10.1109/TBME.2013.2269766.
[11] Stoica P, Babu P. Model order estimation via penalizing adaptively the likelihood(PAL)[J]. Signal Processing, 2013, 93(11): 2865-2871. DOI: 10.1016/j.sigpro.2013.03.014.
[12] Lu S, Ju K H, Chon K H. A new algorithm for linear and nonlinear ARMA model parameter estimation using affine geometry[J]. IEEE Transactions on Biomedical Engineering, 2001, 48(10): 1116-1124. DOI: 10.1109/ 10.951514.
[13] Yang C, Le Bouquin Jeannes R, Bellanger J J, et al. A new strategy for model order identification and its application to transfer entropy for EEG signals analysis[J]. IEEE Transactions on Biomedical Engineering, 2013, 60(5): 1318-1327. DOI: 10.1109/TBME.2012.2234125.

备注/Memo

备注/Memo:
收稿日期: 2017-12-05.
作者简介: 杨淳沨(1981—),男,博士,讲师,chunfeng.yang@seu.edu.cn.
基金项目: 国家自然科学基金资助项目(31400842,31640028,61201344,61271312,61401085)、江苏省自然科学基金资助项目(BK20150650)、教育部留学回国人员科研启动基金资助项目、中国科学院声学所声场声信息国家重点实验室开放课题资助项目(SKLA201604).
引用本文: 杨淳沨,吴国成,伍家松,等.基于gAIC和滑动窗的自回归模型参数估计算法[J].东南大学学报(自然科学版),2018,48(3):381-386. DOI:10.3969/j.issn.1001-0505.2018.03.001.
更新日期/Last Update: 2018-05-20