Time Series Components Separation Based on Singular Spectral Analysis Visualization: an HJ-biplot Method Application

DOI: https://doi.org/10.19139/soic-2310-5070-897
Keywords: Singular Spectrum Analysis, NIPALS algorithm, Biplots

Abstract

The extraction of essential features of any real-valued time series is crucial for exploring, modeling and producing, for example, forecasts. Taking advantage of the representation of a time series data by its trajectory matrix of Hankel constructed using Singular Spectrum Analysis, as well as of its decomposition through Principal Component Analysis via Partial Least Squares, we implement a graphical display employing the biplot methodology. A diversity of types of biplots can be constructed depending on the two matrices considered in the factorization of the trajectory matrix. In this work, we discuss the called HJ-biplot which yields a simultaneous representation of both rows and columns of the matrix with maximum quality. Interpretation of this type of biplot on Hankel related trajectory matrices is discussed from a real-world data set.

References

T. Alexandrov, A method of trend extraction using Singular Spectrum Analysis, REVSTAT, Statistical Journal, vol. 7, n. 1, pp. 1-22, 2009.

K. Gabriel, The biplot graphic display of matrices with application to principal component analysis, Biometrika, vol. 58, n. 3, pp. 453-467, 1971

M.P. Galindo, An alternative of simultaneous representation: HJ-biplot, Questii, vol. 10, 1, pp. 13-23, 1986.

P. Geladi and B.R. Kowalsky, Partial Least Squares regression: a tutorial, Analytica Chimica Acta, vol. 185, pp. 1-17, 1986.

N. Golyandina, V. Korobeynikov and A. Zhigljavsky, Singular Spectrum Analysis with R, 1st ed., Springer, Berlin, 2018.

N. Golyandina, V. Nekrutkin and A. Zhigljavsky, Analysis of Time Series Structure: SSA and Related Techniques, 1st ed. Chapman & Hall/CRC, Boca Raton, Florida, 2001.

N. Golyandina and A. Shlemov, Variations of Singular Spectrum Analysis for separability improvement: non-orthogonal decompositions of time series, Statistics and its Interface, vol. 8, n. 3, pp. 277-294, 2015.

M. Greenacre, Biplots in Practice, FBBVA, Bilbao, Biscay, 2010.

A.B. Nieto, M.P. Galindo, V. Leiva and P.V. Galindo, A methodology for biplots based on bootstrapping with R, Colombian Journal of Statistics, vol. 37, n. 2, pp. 367-397, 2014.

S. Moritz, T. Bartz-Beielstein, imputeTS: Time Series Missing Value Imputation in R, The R Journal, vol. 9, n. 1, pp. 207-218, 2017. https://doi.org/10.32614/RJ-2017-009.

NOAA Homepage, https://www.esrl.noaa.gov/gmd/ccgg/trends/, last accessed 24/05/2019.

R Core Team, R: A language and environment for statistical computing, R Foundation for Statistical Computing, Vienna, Austria, 2019 https://www.R-project.org/.

V.E. Vinzi and G. Russolillo, Partial Least Squares algorithms and methods, WIREs Comput Stat, vol. 5, pp. 1-19, 2013.

H. Wold, Estimation of principal components and related models by iterative least squares, in Multivariate Analysis, edited by P.R. Krishnaiah, Academic Press, New York, pp. 391–420, 1966.

S. Wold, C. Albano, W.J. Dunn, K. Esbensen, S. Hellberg, E. Johansson and M. Sjostrom, Pattern recognition: finding and using regularities in multivariate data, in Food Research and Data Analysis, edited by H. Martens and H. Russwurm, Applied Science Publishers, London, pp. 147-189, 1983.

K. Wright, nipals: Principal Components Analysis using NIPALS or Weighted EMPCA, with Gram-Schmidt Orthogonalization. R package version 0.7, 2020. https://CRAN.R-project.org/package=nipals

Published
2020-05-18
How to Cite
Oliveira da Silva, A., & Freitas, A. (2020). Time Series Components Separation Based on Singular Spectral Analysis Visualization: an HJ-biplot Method Application. Statistics, Optimization & Information Computing, 8(2), 346-358. https://doi.org/10.19139/soic-2310-5070-897
Issue
Vol 8 No 2 (2020)
Section
Research Articles
Authors who publish with this journal agree to the following terms:

    1. Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
    2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
    3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).