Test your surrogate data before you test for nonlinearity

Abstract

The schemes for the generation of surrogate data in order to test the null hypothesis of linear stochastic process undergoing nonlinear static transform are investigated as to their consistency in representing the null hypothesis. In particular, we pinpoint some important caveats of the prominent algorithm of amplitude adjusted Fourier transform surrogates (AAFT) and compare it to the iterated AAFT (IAAFT), which is more consistent in representing the null hypothesis. It turns out that in many applications with real data the inferences of nonlinearity after marginal rejection of the null hypothesis were premature and have to be re-investigated taken into account the inaccuracies in the AAFT algorithm, mainly concerning the mismatching of the linear correlations. In order to deal with such inaccuracies we propose the use of linear together with nonlinear polynomials as discriminating statistics. The application of this setup to some well-known real data sets cautions against the use of the AAFT algorithm.

14 pages, 15 figures, submitted to Physical Review E

References in corpus (3)

• Improved surrogate data for nonlinearity tests
• Interdisciplinary application of nonlinear time series methods
• Constrained randomization of time series data

Cited by in corpus (14)

• Surrogate time series
• A non subjective approach to the GP algorithm for analysing noisy time series
• Statically Transformed Autoregressive Process and Surrogate Data Test for Nonlinearity
• Nonlinear time series anaysis of the light curves from the black hole system GRS1915+105
• Testing for intrinsic multifractality in the global grain spot market indices: A multifractal detrended fluctuation analysis
• Recurrence network measures for hypothesis testing using surrogate data: application to black hole light curves
• Search for chaos in neutron star systems: Is Cyg X-3 a black hole?
• Non-parametric power-law surrogates
• Approximate discrete dynamics of EMG signal
• Accounting for outliers and calendar effects in surrogate simulations of stock return sequences
• Phase synchronization of instrumental music signals
• Combining the ApEn statistic with surrogate data analysis for the detection of nonlinear dynamics in time series
• Scaling behavior in the heart rate variability characteristics with age
• Characterization of Cardio signals by time-frequency domain analysis