Summary
International Symposium on Nonlinear Theory and Its Applications
2015
Session Number:A3L-A
Session:
Number:A3L-A-2
Measuring the Complexity of Time Series Using Ordinal Partition Networks
Michael McCullough, Michael Small, Herbert Ho-Ching Iu,
pp.117-120
Publication Date:2015/12/1
Online ISSN:2188-5079
DOI:10.34385/proc.47.A3L-A-2
PDF download (2.9MB)
Summary:
Recent investigations of ordinal partition networks show potential for the method as a tool for analysing nonlinear time series from continuous systems. In this paper we demonstrate how interpolation can be used to minimise node aliasing such that ordinal partition networks more accurately capture dynamics from discrete time sampled data. We then show how the transitional complexity of a time series can be quantified using a weighted average of node entropy and investigate this measure as applied to the Rossler system.