Intelligent Identification of Trend Components in Singular Spectrum Analysis

被引:3
|
作者
Golyandina, Nina [1 ]
Dudnik, Pavel [1 ]
Shlemov, Alex [1 ]
机构
[1] St Petersburg State Univ, Fac Math & Mech, Univ Skaya Nab 7-9, St Petersburg 199034, Russia
关键词
singular spectrum analysis; time series decomposition; trend extraction; seasonality; automated SSA; ALGORITHM;
D O I
10.3390/a16070353
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Singular spectrum analysis (SSA) is a non-parametric adaptive technique used for time series analysis. It allows solving various problems related to time series without the need to define a model. In this study, we focus on the problem of trend extraction. To extract trends using SSA, a grouping of elementary components is required. However, automating this process is challenging due to the nonparametric nature of SSA. Although there are some known approaches to automated grouping in SSA, they do not work well when the signal components are mixed. In this paper, a novel approach that combines automated identification of trend components with separability improvement is proposed. We also consider a new method called EOSSA for separability improvement, along with other known methods. The automated modifications are numerically compared and applied to real-life time series. The proposed approach demonstrated its advantage in extracting trends when dealing with mixed signal components. The separability-improving method EOSSA proved to be the most accurate when the signal rank is properly detected or slightly exceeded. The automated SSA was very successfully applied to US Unemployment data to separate an annual trend from seasonal effects. The proposed approach has shown its capability to automatically extract trends without the need to determine their parametric form.
引用
收藏
页数:22
相关论文
共 50 条
  • [21] Modeling the Trend of Credit Card Usage Behavior for Different Age Groups Based on Singular Spectrum Analysis
    Nai, Wei
    Liu, Lu
    Wang, Shaoyin
    Dong, Decun
    ALGORITHMS, 2018, 11 (02):
  • [22] Functional singular spectrum analysis
    Haghbin, Hossein
    Najibi, Seyed Morteza
    Mahmoudvand, Rahim
    Trinka, Jordan
    Maadooliat, Mehdi
    STAT, 2021, 10 (01):
  • [23] Nonlinear singular spectrum analysis
    Hsieh, WW
    Wu, AM
    PROCEEDING OF THE 2002 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOLS 1-3, 2002, : 2819 - 2824
  • [24] Trend forecasting based on Singular Spectrum Analysis of traffic workload in a large-scale wireless LAN
    Tzagkarakis, George
    Papadopouli, Maria
    Tsakalides, Panagiotis
    PERFORMANCE EVALUATION, 2009, 66 (3-5) : 173 - 190
  • [25] Posterior Singular Spectrum Analysis
    Holmstrom, Lasse
    Launonen, Ilkka
    STATISTICAL ANALYSIS AND DATA MINING, 2013, 6 (05) : 387 - 402
  • [26] Analysis of the experimental absorption spectrum of the rabbit lung and identification of its components
    Pinheiro, Maria R.
    Tuchin, Valery V.
    Oliveira, Luis M.
    JOURNAL OF BIOPHOTONICS, 2024,
  • [27] Analysis of the experimental absorption spectrum of the rabbit lung and identification of its components
    Pinheiro, Maria R.
    Tuchin, Valery V.
    Oliveira, Luís M.
    Journal of Biophotonics, 2024,
  • [28] ANALYSIS OF THE COMPONENTS OF A LINEAR TREND IN PROPORTIONS
    PEARCE, NE
    CRYER, PC
    AMERICAN JOURNAL OF EPIDEMIOLOGY, 1986, 124 (01) : 127 - 133
  • [29] Grouping and Selecting Singular Spectrum Analysis Components for Denoising Via Empirical Mode Decomposition Approach
    Peiru Lin
    Weichao Kuang
    Yuwei Liu
    Bingo Wing-Kuen Ling
    Circuits, Systems, and Signal Processing, 2019, 38 : 356 - 370
  • [30] A principal components/singular spectrum analysis approach to ENSO and PDO influences on rainfall in western Iran
    Biabanaki, Monireh
    Eslamian, Seyed Saeid
    Koupai, Jahangir Abedi
    Canon, Julio
    Boni, Giorgio
    Gheysari, Mandi
    HYDROLOGY RESEARCH, 2014, 45 (02): : 250 - 262