×

zbMATH — the first resource for mathematics

Nonlinear Laplacian spectral analysis for time series with intermittency and low-frequency variability. (English) Zbl 1256.62053
Summary: Many processes in science and engineering develop multiscale temporal and spatial patterns, with complex underlying dynamics and time-dependent external forcings. Because of the importance in understanding and predicting these phenomena, extracting the salient modes of variability empirically from incomplete observations is a problem of wide contemporary interest. We present a technique for analyzing high-dimensional, complex time series that exploits the geometrical relationships between the observed data points to recover features characteristic of strongly nonlinear dynamics (such as intermittency and rare events), which are not accessible to classical singular spectrum analysis. The method employs Laplacian eigenmaps, evaluated after suitable time-lagged embedding, to produce a reduced representation of the observed samples, where standard tools of matrix algebra can be used to perform truncated singular-value decompositions despite the nonlinear geometrical structure of the data set. We illustrate the utility of the technique in capturing intermittent modes associated with the Kuroshio current in the North Pacific sector of a general circulation model and dimensional reduction of a low-order atmospheric model featuring chaotic intermittent regime transitions, where classical singular spectrum analysis is already known to fail dramatically.

MSC:
62M15 Inference from stochastic processes and spectral analysis
62M10 Time series, auto-correlation, regression, etc. in statistics (GARCH)
86A32 Geostatistics
62P12 Applications of statistics to environmental and related topics
PDF BibTeX Cite
Full Text: DOI
References:
[1] Sreenivasan, Annual Review of Fluid Mechanics 29 (1) pp 435– (1997)
[2] PHYS D 35 pp 395– (1989) · Zbl 0709.62628
[3] J ATMOS SCI 58 pp 1341– (2001)
[4] J ATMOS SCI 61 pp 235– (2004)
[5] LECT NOTES COMP SCI ENG 4 pp 98– (1998)
[6] 21 pp 113– (2006) · Zbl 1103.60069
[7] Astrophysical Journal 566 pp 276– (2002)
[8] SOLAR PHYS 248 pp 247– (2008)
[9] 64 pp 683– (1991) · Zbl 0943.37510
[10] REV GEOPHYS 40 pp 1003– (2002)
[11] DYNAMICAL SYSTEMS AND TURBULENCE WARWICK LECTURE NOTES IN MATHEMATICS Vol. 898 pp 366– (1981)
[12] 65 pp 579– (1991) · Zbl 0943.37506
[13] Deyle 6 (3) pp e18295– (2011)
[14] PHYS D 20 pp 217– (1986) · Zbl 0603.58040
[15] 14 pp 483– (1993) · Zbl 0774.65084
[16] J ATMOS SCI 61 pp 2206– (2004)
[17] Belkin, Neural Computation 15 (6) pp 1373– (2003) · Zbl 1085.68119
[18] 21 pp 5– (2006) · Zbl 1095.68094
[19] J CLIMATE 19 pp 2496– (2006)
[20] CLIMATE DYN 36 pp 1813– (2010)
[21] J CLIMATE 23 pp 6292– (2010)
[22] J OCEANOGR 58 pp 35– (2002)
[23] PHYS D 36 pp 222– (1989) · Zbl 0687.76045
[24] J ATMOS SCI 61 pp 1406– (2004)
[25] Physical Review Letters 45 pp 712– (1980)
[26] COMMUN MATH SCI 8 pp 145– (2010) · Zbl 1201.37109
[27] J ATMOS SCI 65 pp 1941– (2008)
[28] PNAS 102 (21) pp 7426– (2005)
[29] ADV NIPS 15 pp 929– (2003)
[30] J COMPUT SYST SCI 74 pp 1289– (2008) · Zbl 1157.68056
[31] PNAS 105 (6) pp 1803– (2008) · Zbl 1215.58012
[32] Hsieh, Neural networks : the official journal of the International Neural Network Society 20 (4) pp 434– (2007) · Zbl 1115.68129
[33] PHYS REV E 55 pp 5365– (1997)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.