Display Abstract

Title Order patterns: a diagnostic tool for time series

Name Christoph Bandt
Country Germany
Email bandt@uni-greifswald.de
Co-Author(s) Christoph Bandt
Submit Time 2014-02-21 06:06:17
Session
Special Session 68: Entropy-like quantities and applications
Contents
Permutation entropy is widely used to evaluate the complexity of time series of moderate length. For long time series, multiscale analysis of all order patterns will give more information, as we shall demonstrate with examples from medicine, wheather, internet traffic, and music. A very interesting case is pink noise where frequencies of order patterns do not depend on the scale, We analyse several models and applications of pink noise and present a simple new model.