Detecting and analysing nonstationarity in a time series with nonlinear cross-predictions

Abstract

We propose an informal test for stationarity in a time series which checks for the compatibility of nonlinear approximations to the dynamics made in different segments of the sequence. The segments are compared directly, rather than via statistical parameters. The approach provides detailed information about episodes with similar dynamics during the measurement period. Thus physically relevant changes in the dynamics can be followed.

4 pages, 4 figures

Cited by in corpus (7)

• Scale Invariance in the Nonstationarity of Physiological Signals
• Time series analysis for minority game simulations of financial markets
• Time-series-analysis-based detection of critical transitions in real-world non-autonomous systems
• The dynamics of laser droplet generation
• Fluctuation of similarity (FLUS) to detect transitions between distinct dynamical regimes in short time series
• Low dimensional behavior in three-dimensional coupled map lattices
• Divergence Measure Between Chaotic Attractors