Circulant singular spectrum analysis: A new automated procedure for signal extraction

Sometimes, it is of interest to single out the fluctuations associated to a given frequency. We propose a new variant of SSA, Circulant SSA (CiSSA), that allows to extract the signal associated to any frequency specified beforehand. This is a novelty when compared with other procedures that need to identify ex-post the frequencies associated to extracted signals. We prove that CiSSA is asymptotically equivalent to these alternative procedures although with the advantage of avoiding the need of the subsequent frequency identification. We check its good performance and compare it to alternative SSA methods through several simulations for linear and nonlinear time series. We also prove its validity in the nonstationary case. To show how it works with real data, we apply CiSSA to extract the business cycle and deseasonalize the Industrial Production Index of six countries. Economists follow this indicator in order to assess the state of the economy in real time. We find that the estimated cycles match the dated recessions from the OECD showing its reliability for business cycle analysis. Finally, we analyze the strong separability of the estimated components. In particular, we check that the deseasonalized time series do not show any evidence of residual seasonality.

[1]  J. Burman.  Seasonal Adjustment by Signal Extraction , 1980 .

[2]  Robert M. Gray,et al.  On the asymptotic eigenvalue distribution of Toeplitz matrices , 1972, IEEE Trans. Inf. Theory.

[3]  Dimitrios D. Thomakos,et al.  A review on singular spectrum analysis for economic and financial time series , 2010 .

[4]  Enrico Bozzo,et al.  Relationship between Singular Spectrum Analysis and Fourier analysis: Theory and application to the monitoring of volcanic activity , 2010, Comput. Math. Appl..

[5]  Massimo Bilancia,et al.  Airborne Particulate Matter and Adverse Health Events: Robust Estimation of Timescale Effects , 2010 .

[6]  J. Arteche,et al.  Singular Spectrum Analysis for signal extraction in Stochastic Volatility models , 2017 .

[7]  Rahim Mahmoudvand,et al.  Missing value imputation in time series using Singular Spectrum Analysis , 2016 .

[8]  Nina Golyandina.  Statistical approach to detection of signals by Monte Carlo singular spectrum analysis: Multiple testing , 2019 .

[9]  P. Tilli.  Singular values and eigenvalues of non-hermitian block Toeplitz matrices , 1996 .

[10]  Soo-Chang Pei,et al.  The Mystery Curve: A Signal Processing Point of View [Lecture Notes] , 2017, IEEE Signal Processing Magazine.

[11]  Manuel Duarte Ortigueira,et al.  On the HHT, its problems, and some solutions , 2008 .

[12]  N. Huang,et al.  The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis , 1998, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[13]  Anatoly Zhigljavsky,et al.  Forecasting UK Industrial Production with Multivariate Singular Spectrum Analysis , 2013 .

[14]  Klaus Fraedrich,et al.  Estimating the Dimensions of Weather and Climate Attractors , 1986 .

[15]  V. Pisarenko.  The Retrieval of Harmonics from a Covariance Function , 1973 .

[16]  Mohammad Pooyan,et al.  Improving the performance of the SSVEP-based BCI system using optimized singular spectrum analysis (OSSA) , 2018, Biomed. Signal Process. Control..

[17]  Emmanuel Sirimal Silva,et al.  On the use of singular spectrum analysis for forecasting U.S. trade before, during and after the 2008 recession , 2015 .

[18]  Andrew L. Rukhin,et al.  Analysis of Time Series Structure SSA and Related Techniques , 2002, Technometrics.

[19]  Pedro M. Crespo,et al.  Block Toeplitz Matrices: Asymptotic Results and Applications , 2012, Found. Trends Commun. Inf. Theory.

[20]  Norden E. Huang,et al.  Ensemble Empirical Mode Decomposition: a Noise-Assisted Data Analysis Method , 2009, Adv. Data Sci. Adapt. Anal..

[21]  W. F. Trench.  Absolute equal distribution of the spectra of Hermitian matrices , 2003 .

[22]  Anton Korobeynikov,et al.  Computation- and Space-Efficient Implementation of SSA , 2009, ArXiv.

[23]  Giorgio Biagetti,et al.  Analysis of the EMG Signal During Cyclic Movements Using Multicomponent AM–FM Decomposition , 2015, IEEE Journal of Biomedical and Health Informatics.

[24]  Abdol S. Soofi,et al.  Predicting inflation dynamics with singular spectrum analysis , 2013 .

[25]  Leonard A. Smith,et al.  Monte Carlo SSA: Detecting irregular oscillations in the Presence of Colored Noise , 1996 .

[26]  R. Vautard,et al.  Singular spectrum analysis in nonlinear dynamics, with applications to paleoclimatic time series , 1989 .

[27]  Donald Poskitt,et al.  Forecasting stochastic processes using singular spectrum analysis: Aspects of the theory and application , 2017 .

[28]  KENJI KUME,et al.  Interpretation of singular Spectrum Analysis as Complete eigenfilter Decomposition , 2012, Adv. Data Sci. Adapt. Anal..

[29]  G. P. King,et al.  Extracting qualitative dynamics from experimental data , 1986 .

[30]  Gary Brown,et al.  Forecasting before, during, and after recession with singular spectrum analysis , 2013 .

[31]  Michael Ghil,et al.  ADVANCED SPECTRAL METHODS FOR CLIMATIC TIME SERIES , 2002 .

[32]  Miguel de Carvalho,et al.  Real-Time Nowcasting the US Output Gap: Singular Spectrum Analysis at Work , 2017 .

[33]  S. S. Shen,et al.  Applications of Hilbert–Huang transform to non‐stationary financial time series analysis , 2003 .

[34]  U. Grenander,et al.  Toeplitz Forms And Their Applications , 1958 .

[35]  Michael Feldman,et al.  Analytical basics of the EMD: Two harmonics decomposition , 2009 .

[36]  Nina Golyandina,et al.  Particularities and commonalities of singular spectrum analysis as a method of time series analysis and signal processing , 2019, WIREs Computational Statistics.

[37]  J. Smith,et al.  FAST MUSIC – AN EFFICIENT IMPLEMENTATION OF THE MUSIC ALGORITHM FOR FREQUENCY ESTIMATION OF APPROXIMATELY PERIODIC SIGNALS , 2018 .

[38]  James Durbin,et al.  Time Series Analysis by State Space Methods: Second Edition , 2012 .

[39]  Miguel de Carvalho,et al.  Tracking the US business cycle with a singular spectrum analysis , 2012 .

[40]  Ana Maria Tomé,et al.  On the use of Singular Spectrum Analysis. , 2018, 1807.10679.

[41]  Nina Golyandina,et al.  Automatic extraction and forecast of time series cyclic components within the framework of SSA , 2005 .

[42]  Gang Liu,et al.  A novel blind source separation method for single-channel signal , 2010, Signal Process..

[43]  Giorgio Biagetti,et al.  Multicomponent AM–FM Representations: An Asymptotically Exact Approach , 2007, IEEE Transactions on Audio, Speech, and Language Processing.

[44]  Brent R. Moulton.  Residual Seasonality in GDP and GDI: Findings and Next Steps , 2016 .

[45]  Nina Golyandina,et al.  Basic Singular Spectrum Analysis and forecasting with R , 2012, Comput. Stat. Data Anal..

[46]  Anatoly A. Zhigljavsky,et al.  Singular Spectrum Analysis for Time Series , 2013, International Encyclopedia of Statistical Science.

[47]  Michael Ghil,et al.  Economic Cycles and Their Synchronization: A Comparison of Cyclic Modes in Three European Countries , 2016 .

[48]  Peter C. Young,et al.  Recursive Estimation and Time-Series Analysis: An Introduction , 1984 .

[49]  Michael Ghil,et al.  Intraseasonal oscillations in the global atmosphere. I - Northern Hemisphere and tropics , 1991 .

[50]  D. R. Salgado,et al.  Analysis of the structure of vibration signals for tool wear detection , 2008 .

[51]  Robert M. Gray,et al.  On Unbounded Toeplitz Matrices and Nonstationary Time Series with an Application to Information Theory , 1974, Inf. Control..

[52]  Siem Jan Koopman,et al.  Time Series Analysis by State Space Methods , 2001 .

[53]  Lars Eldén,et al.  Fast computation of the principal singular vectors of Toeplitz matrices arising in exponential data modelling , 1996, Signal Process..

[54]  Peter Lancaster,et al.  The theory of matrices , 1969 .

[55]  M. S. Solary.  Finding eigenvalues for heptadiagonal symmetric Toeplitz matrices , 2013 .

[56]  R. Vautard,et al.  Singular-spectrum analysis: a toolkit for short, noisy chaotic signals , 1992 .

[57]  F. Gianfelici,et al.  AM-FM decomposition of speech signals: an asymptotically exact approach based on the iterated hilbert transform , 2005, IEEE/SP 13th Workshop on Statistical Signal Processing, 2005.

[58]  Michael Feldman,et al.  Time-varying vibration decomposition and analysis based on the Hilbert transform , 2006 .

[59]  Estela Bee Dagum,et al.  Modelling, Forecasting and Seasonally Adjusting Economic Time Series with the X-11 ARIMA Method , 1978 .

[60]  T. Alexandrov,et al.  Automatic extraction and forecast of time series cyclic components within the framework of SSA , 2005 .

[61]  Judea Pearl,et al.  On coding and filtering stationary signals by discrete Fourier transforms (Corresp.) , 1973, IEEE Trans. Inf. Theory.

[62]  Manuel Duarte Ortigueira,et al.  Eigendecomposition versus singular value decomposition in adaptive array signal processing , 1991, Signal Process..

[63]  A. Yurova,et al.  Atmospheric heat advection in the Kara Sea region under main synoptic processes , 2018, International Journal of Climatology.

[64]  Ujjwal Kumar,et al.  Time series models (Grey-Markov, Grey Model with rolling mechanism and singular spectrum analysis) to forecast energy consumption in India , 2010 .

[65]  Salim Lahmiri,et al.  Minute-ahead stock price forecasting based on singular spectrum analysis and support vector regression , 2018, Appl. Math. Comput..