Statistical Test for Dynamical Nonstationarity in Observed Time-Series Data

1995/12/21 by Matthew B. Kennel, Kennel, Matthew B.
Physics and Astronomy · #Chaotic Dynamics (nlin.CD) #FOS: Physical sciences #chao-dyn #nlin.CD

paper · pdf · doi:10.48550/arxiv.chao-dyn/9512005

REVTEX, 4 encapsulated postscript figures, in uuencoded, gzipped tar archive; email: [email protected]

arxiv created 1995/12/21 · arxiv updated 2009/11/30

Abstract

Information in the time distribution of points in a state space reconstructed from observed data yields a test for ``nonstationarity''. Framed in terms of a statistical hypothesis test, this numerical algorithm can discern whether some underlying slow changes in parameters have taken place. The method examines a fundamental object in nonlinear dynamics, the geometry of orbits in state space, with corrections to overcome difficulties in real dynamical data which cause naive statistics to fail.

Related

• A TEST FOR STATIONARITY: FINDING PARTS IN TIME SERIES APT FOR CORRELATION DIMENSION ESTIMATES
• A Method for Detecting Possible Non-determinism in a Time Series
• Nonstationary open dynamical systems
• Tests for partial correlation between repeatedly observed nonstationary nonlinear timeseries
• Nonparametric Multiple Change Point Detection for Non-Stationary Times Series