Revealing mutual influence of oscillatory systems from the observation data

In different fields of physical research, we often meet the problem of distinguishing between bidirectional and unidirectional couplings of oscillatory systems, i.e., mutual influence and one-way influence, on the basis of the observed time series. In this work, it is shown that a widely used approa...

Full description

Saved in:
Bibliographic Details
Published in:Radiophysics and quantum electronics 2013-03, Vol.55 (10-11), p.662-675
Main Authors: Smirnov, D. A., Bezruchko, B. P.
Format: Article
Language:English
Subjects:
Astronomy
Astrophysics and Astroparticles
Bidirectional
Confidence intervals
Couplings
Hadrons
Heavy Ions
Joining
Lasers
Mathematical and Computational Physics
Mathematical models
Nuclear Physics
Observations and Techniques
Optical Devices
Optics
Photonics
Physics
Physics and Astronomy
Quantum Optics
Sampling
Statistical tests
Theoretical
Time series
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In different fields of physical research, we often meet the problem of distinguishing between bidirectional and unidirectional couplings of oscillatory systems, i.e., mutual influence and one-way influence, on the basis of the observed time series. In this work, it is shown that a widely used approach, based on the calculation of prediction errors of empirical models (Granger causality), may give spurious conclusions on bidirectional coupling if a sampling interval exceeds a certain intrinsic time scale of the studied systems. Taking this effect into account, we propose and illustrate by standard examples a statistical test which allows one to reveal bidirectional coupling with specified confidence probability.
ISSN:0033-8443
1573-9120
DOI:10.1007/s11141-013-9404-6