Re-evaluating the performance of the nonlinear prediction error for the detection of deterministic dynamics

The nonlinear prediction error in combination with appropriately generated surrogate data is widely used as a discriminating statistic for weak nonlinearity. In this paper two of the most widely used formulations of this particular measure are evaluated using data from exemplary nonlinear and linear systems. It is found that both methods, contrary to expectation and accepted utility, give rise to substantial false positive and false negative detection rates in purely linear and nonlinear systems respectively. These findings suggest that the nonlinear prediction error should be used with considerable caution. The implications of these findings for the search for nonlinearity in the human electroencephalogram are briefly discussed.

[1]  D. Kugiumtzis,et al.  Test your surrogate data before you test for nonlinearity. , 1999, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[2]  T. Schreiber,et al.  Discrimination power of measures for nonlinearity in a time series , 1997, chao-dyn/9909043.

[3]  Dimitris Kugiumtzis,et al.  Evaluation of Surrogate and Bootstrap Tests for Nonlinearity in Time Series , 2008 .

[4]  Danilo P Mandic,et al.  Indications of nonlinear structures in brain electrical activity. , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.

[5]  Ernst Fernando Lopes Da Silva Niedermeyer,et al.  Electroencephalography, basic principles, clinical applications, and related fields , 1982 .

[6]  K Lehnertz,et al.  Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity: dependence on recording region and brain state. , 2001, Physical review. E, Statistical, nonlinear, and soft matter physics.

[7]  M. Small.  Applied Nonlinear Time Series Analysis: Applications in Physics, Physiology and Finance , 2005 .

[8]  E. Růžička,et al.  Investigation of non-linear properties of multichannel EEG in the early stages of Parkinson's disease , 2001, Clinical Neurophysiology.

[9]  Holger Kantz,et al.  Practical implementation of nonlinear time series methods: The TISEAN package. , 1998, Chaos.

[10]  Jens Haueisen,et al.  Alpha entrainment in human electroencephalogram and magnetoencephalogram recordings , 2006, Neuroreport.

[11]  J. S. Barlow,et al.  Methods of Analysis of Nonstationary EEGs, with Emphasis on Segmentation Techniques: A Comparative Review , 1985, Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society.

[12]  W J Levy,et al.  Effect of Epoch Length on Power Spectrum Analysis of the EEG , 1987, Anesthesiology.

[13]  A. Fingelkurts,et al.  Morphology and dynamic repertoire of EEG short-term spectral patterns in rest: Explorative study , 2010, Neuroscience Research.

[14]  P. A. Robinson,et al.  Automated characterization of multiple alpha peaks in multi-site electroencephalograms , 2008, Journal of Neuroscience Methods.

[15]  F. L. D. Silva,et al.  Dynamics of the human alpha rhythm: evidence for non-linearity? , 1999, Clinical Neurophysiology.

[16]  T. Schreiber,et al.  Surrogate time series , 1999, chao-dyn/9909037.

[17]  C. Stam,et al.  Small‐world properties of nonlinear brain activity in schizophrenia , 2009, Human brain mapping.

[18]  Anastasios A. Tsonis,et al.  Reconstructing Dynamics from Observables: the Issue of the Delay Parameter Revisited , 2007, Int. J. Bifurc. Chaos.

[19]  A. Fingelkurts,et al.  Operational Architectonics of the Human Brain Biopotential Field: Towards Solving the Mind-Brain Problem , 2001 .

[20]  J. Sprott.  Chaos and time-series analysis , 2001 .

[21]  T Inouye,et al.  A new segmentation method of electroencephalograms by use of Akaike's information criterion. , 1995, Brain research. Cognitive brain research.

[22]  M. Karlický,et al.  Analysis of solar narrow band dm-spikes observed at 1420 and 2695 MHz , 2003 .

[23]  Piet M. T. Broersen,et al.  Automatic spectral analysis with time series models , 2002, IEEE Trans. Instrum. Meas..

[24]  Bard Ermentrout,et al.  Simulating, analyzing, and animating dynamical systems - a guide to XPPAUT for researchers and students , 2002, Software, environments, tools.

[25]  Tanya Schmah,et al.  Surrogate Data Pathologies and the False-positive rejection of the Null Hypothesis , 2001, Int. J. Bifurc. Chaos.

[26]  H. Kantz,et al.  Nonlinear time series analysis , 1997 .

[27]  G. B. Anderson,et al.  Modeling the Stationarity and Gaussianity of Spontaneous Electroencephalographic Activity , 1975, IEEE Transactions on Biomedical Engineering.

[28]  George Sugihara,et al.  Nonlinear forecasting as a way of distinguishing chaos from measurement error in time series , 1990, Nature.

[29]  Patrick J O Miller,et al.  Nonlinear phenomena in the vocalizations of North Atlantic right whales (Eubalaena glacialis) and killer whales (Orcinus orca). , 2007, The Journal of the Acoustical Society of America.

[30]  A. Kraskov,et al.  Bivariate surrogate techniques: necessity, strengths, and caveats. , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.

[31]  Tohru Ikeguchi,et al.  Effects of data windows on the methods of surrogate data. , 2005, Physical review. E, Statistical, nonlinear, and soft matter physics.