Summary

Proceedings of the 2012 International Symposium on Nonlinear Theory and its Applications

2012

Session Number:B2L-C

Session:

Number:373

Verifying chaotic dynamics from experimental data

Michael Small,  David M. Walker,  Antoinette Tordesillas,  

pp.373-376

Publication Date:

Online ISSN:2188-5079

DOI:10.15248/proc.1.373

PDF download (559.7KB)

Often, one is faced with measured time series data from some (presumed to be deterministic) dynamical system. The problem is to correctly infer the true, or at least, likely, underlying dynamical system from data alone. A variety of methods exist to achieve this — under the general umbrella on nonlinear modeling and machine learning. These methods fit a surface (usually smooth) to the data in such a way that that surface can be used as a proxy for the evolution operator of the original system. Unfortunately, different methods produce different results. Worse still, due to the nonlinearity inherent to the problem, even the same method will produce a range of distinct local minima. The aim of this report is to apply an ensemble of dynamical measures of system behaviour to show how one can determine which models behave most like the underlying data.

[1] C. Diks. Estimating invariants of noisy attractors. Physical Review E, 53:R4263-R4266, 1996.

[2] K. Judd and A. Mees. On selecting models for nonlinear time series. Physica D, 82:426-444, 1995.

[3] J. Maquet, C. Letellier, and L. A. Aguirre. Global models from the canadian lynx cycles as a direct evidence for chaos in real ecosystems. J. Math. Biol., 55:21-39, 2007.

[4] M. Small. Applied Nonlinear Time Series Analysis: Applications in Physics, Physiology and Finance, volume 52 of Nonlinear Science Series A. World Scientific, Singapore, 2005.

[5] M. Small and C. Carmeli. Re-examination of evidence for low-dimensional chaos in the canadian lynx data. In International Symposium on Nonlinear Theory and its Applications. Research Society of Nonlinear Theory and its Applications, IEICE, 2009.

[6] M. Small and K. Judd. Correlation dimension: A pivotal statistic for non-constrained realizations of composite hypotheses in surrogate data analysis. Physica D, 120:386-400, 1998.

[7] F. Takens. Detecting strange attractors in turbulence. Lecture Notes in Mathematics, 898:366-381, 1981.

[8] A. Tordesillas. Force chain buckling, unjaming transition and shear banding in dense granular assemblies. Phil. Mag., 87(32):4987-5016, 2007.

[9] X.-K. Xu, J. Zhang, and M. Small. Superfamily phenomena and motifs of networks induced from time series. Proc Natl Acad Sci USA, 105:19601-19605, 2008.

[10] D. Yu, M. Small, R. G. Harrison, and C. Diks. Efficient implementation of the Gaussian kernel algorithm in estimating invariants and noise level from noisy time series data. Physical Review E, 61:3750-3756, 2000.