Chin. Phys. Lett.  2017, Vol. 34 Issue (12): 120501    DOI: 10.1088/0256-307X/34/12/120501
GENERAL |
Generalized Multivariate Singular Spectrum Analysis for Nonlinear Time Series De-Noising and Prediction
Yi Ji1, Hong-Bo Xie2**
1School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013
2ARC Centre of Excellence for Mathematical and Statistical Frontiers, Queensland University of Technology, Brisbane, QLD 4000, Australia
Cite this article:   
Yi Ji, Hong-Bo Xie 2017 Chin. Phys. Lett. 34 120501
Download: PDF(927KB)   PDF(mobile)(926KB)   HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks
Abstract Singular spectrum analysis and its multivariate or multichannel singular spectrum analysis (MSSA) variant are effective methods for time series representation, denoising and prediction, with broad application in many fields. However, a key element in MSSA is singular value decomposition of a high-dimensional matrix stack of component matrices, where the spatial (structural) information among multivariate time series is lost or distorted. This vector-space model also leads to difficulties including high dimensionality, small sample size, and numerical instability when applied to multi-dimensional time series. We present a generalized multivariate singular spectrum analysis (GMSSA) method to simultaneously decompose multivariate time series into constituent components, which can overcome the limitations of conventional multivariate singular spectrum analysis. In addition, we propose a SampEn-based method to determine the dominant components in GMSSA. We demonstrate the effectiveness and efficiency of GMSSA to simultaneously de-noise multivariate time series for attractor reconstruction, and to predict both simulated and real-world multivariate noisy time series.
Received: 01 August 2017      Published: 24 November 2017
PACS:  05.45.Tp (Time series analysis)  
  05.45.Ac (Low-dimensional chaos)  
  95.75.Wx (Time series analysis, time variability)  
Fund: Supported by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).
TRENDMD:   
URL:  
https://cpl.iphy.ac.cn/10.1088/0256-307X/34/12/120501       OR      https://cpl.iphy.ac.cn/Y2017/V34/I12/120501
Service
E-mail this article
E-mail Alert
RSS
Articles by authors
Yi Ji
Hong-Bo Xie
[1]Xie H B, Guo T, Sivakumar B, Liew A W C and Dokos S 2014 Proc. R. Soc. A 470 20140409
[2]Hassani H, Heravi S and Zhigljavsky A 2013 J. Forecasting 32 395
[3]Hossein H 2013 Int. J. Energy Stat. 1 55
[4]Song J J, Ren Y and Yan F L 2009 Comput. Biol. Chem. 33 408
[5]Pilgram B and Schappacher W 1998 Int. J. Bifurcation Chaos Appl. Sci. Eng. 8 571
[6]Xie H B, Guo J Y and Zheng Y P 2010 Phys. Lett. A 374 3926
[7]Xie H B, He W X and Liu H 2008 Phys. Lett. A 372 7140
[8]Xie H B, Chen W T, He W X and Liu H 2011 Appl. Soft Comput. 11 2871
[9]Xie H B, Guo J Y and Zheng Y P 2010 Ann. Biomed. Eng. 38 1483
[10]Kennel M B, Brown R and Abarbanel H D I 1992 Phys. Rev. A 45 3403
[11]Jiang J and Xie H B 2016 Chin. Phys. Lett. 33 100501
[12]Wu C L and Chau K W 2013 Eng. Appl. Artif. Intel. 26 997
[13]Wang X Y and Han M 2015 Eng. Appl. Artif. Intel. 40 28
[14]du Preez J and Witt S F 2003 Int. J. Forecasting 19 435
[15]Xie H B, Sivakumar B, Boontra T W and Mengersen K 2017 IEEE Trans. Fuzzy Syst. (in press)
Related articles from Frontiers Journals
[1] Jia-Chen Zhang , Wei-Kai Ren , and Ning-De Jin. Rescaled Range Permutation Entropy: A Method for Quantifying the Dynamical Complexity of Extreme Volatility in Chaotic Time Series[J]. Chin. Phys. Lett., 2020, 37(9): 120501
[2] Ying Du, Jiaqi Liu, Shihui Fu. Information Transmitting and Cognition with a Spiking Neural Network Model[J]. Chin. Phys. Lett., 2018, 35(9): 120501
[3] Sheng-Li Zhu, Lu Gan. Specific Emitter Identification Based on Visibility Graph Entropy[J]. Chin. Phys. Lett., 2018, 35(3): 120501
[4] Sheng-Li Zhu, Lu Gan. Chaos Identification Based on Component Reordering and Visibility Graph[J]. Chin. Phys. Lett., 2017, 34(5): 120501
[5] Ya-Tong Zhou, Yu Fan, Zi-Yi Chen, Jian-Cheng Sun. Multimodality Prediction of Chaotic Time Series with Sparse Hard-Cut EM Learning of the Gaussian Process Mixture Model[J]. Chin. Phys. Lett., 2017, 34(5): 120501
[6] Jian Jiang, Hong-Bo Xie. Denoising Nonlinear Time Series Using Singular Spectrum Analysis and Fuzzy Entropy[J]. Chin. Phys. Lett., 2016, 33(10): 120501
[7] Wei Zheng, Qian Pan, Chen Sun, Yu-Fan Deng, Xiao-Kang Zhao, Zhao Kang. Fractal Analysis of Mobile Social Networks[J]. Chin. Phys. Lett., 2016, 33(03): 120501
[8] ZHANG Ang-Hui, LI Xiao-Wen, SU Gui-Feng, ZHANG Yi. A Multifractal Detrended Fluctuation Analysis of the Ising Financial Markets Model with Small World Topology[J]. Chin. Phys. Lett., 2015, 32(09): 120501
[9] HAO Qing-Yang, JIN Ning-De, HAN Yun-Feng, GAO Zhong-Ke, ZHAI Lu-Sheng. Multi-Scale Time Asymmetry for Detecting the Breakage of Slug Flow Structure[J]. Chin. Phys. Lett., 2014, 31(12): 120501
[10] QU Hua, MA Wen-Tao, ZHAO Ji-Hong, CHEN Ba-Dong. Kernel Least Mean Kurtosis Based Online Chaotic Time Series Prediction[J]. Chin. Phys. Lett., 2013, 30(11): 120501
[11] FAN Chun-Ling, JIN Ning-De, CHEN Xiu-Ting, GAO Zhong-Ke. Multi-Scale Permutation Entropy: A Complexity Measure for Discriminating Two-Phase Flow Dynamics[J]. Chin. Phys. Lett., 2013, 30(9): 120501
[12] ZHANG Xin-Wang, JIN Ning-De, GAO Zhong-Ke, ZHAI Lu-Sheng . Local Property of Recurrence Network for Investigating Gas-Liquid Two-Phase Flow Characteristics[J]. Chin. Phys. Lett., 2013, 30(5): 120501
[13] JIANG Jian, WANG Ru, Pezeril Michel, Wang Qiuping Alexandre. Long Division Unites or Long Union Divides: a Model for Social Network Evolution[J]. Chin. Phys. Lett., 2013, 30(3): 120501
[14] YAN Chuan-Kui, WANG Ru-Bin. Non-identical Neural Network Synchronization Study Based on an Adaptive Learning Rule of Synapses[J]. Chin. Phys. Lett., 2012, 29(9): 120501
[15] QU Jing-Yi, and WANG Ru-Bin. Dynamics of a Cortical Neural Network Based on a Simple Model[J]. Chin. Phys. Lett., 2012, 29(8): 120501
Viewed
Full text


Abstract