📄 surrogates.html
字号:
<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 2.0//EN"><!--Converted with LaTeX2HTML 96.1-h (September 30, 1996) by Nikos Drakos (nikos@cbl.leeds.ac.uk), CBLU, University of Leeds --><HTML><HEAD><TITLE>Surrogate time series</TITLE><META NAME="description" CONTENT="Surrogate time series"><META NAME="keywords" CONTENT="Surrogates"><META NAME="resource-type" CONTENT="document"><META NAME="distribution" CONTENT="global"><LINK REL=STYLESHEET HREF="Surrogates.css"></HEAD><BODY bgcolor=#ffffff LANG="EN" > <A NAME="tex2html29" HREF="node1.html"><IMG WIDTH=37 HEIGHT=24 ALIGN=BOTTOM ALT="next" SRC="next_motif.gif"></A> <IMG WIDTH=26 HEIGHT=24 ALIGN=BOTTOM ALT="up" SRC="up_motif_gr.gif"> <IMG WIDTH=63 HEIGHT=24 ALIGN=BOTTOM ALT="previous" SRC="previous_motif.gif"> <BR><B> Next:</B> <A NAME="tex2html30" HREF="node1.html">Introduction</A><BR> <P><P><H1 ALIGN=CENTER>Surrogate time series</H1><P ALIGN=CENTER><STRONG>Thomas Schreiber and Andreas Schmitz<BR> Physics Department, University of Wuppertal, D-42097 Wuppertal, Germany</STRONG></P><P><P><EM>Abstract</EM> Before we apply nonlinear techniques, for example those inspired by chaos theory, to dynamical phenomena occurring in nature, it is necessary to first ask if the use of such advanced techniques is justified <EM>by the data</EM>. While many processes in nature seem very unlikely a priori to be linear, the possible nonlinear nature might not be evident in specific aspects of their dynamics. The method of surrogate data has become a very popular tool to address such a question. However, while it was meant to provide a statistically rigorous, foolproof framework, some limitations and caveats have shown up in its practical use. In this paper, recent efforts to understand the caveats, avoid the pitfalls, and to overcome some of the limitations, are reviewed and augmented by new material. In particular, we will discuss specific as well as more general approaches to constrained randomisation, providing a full range of examples. New algorithms will be introduced for unevenly sampled and multivariate data and for surrogate spike trains. The main limitation, which lies in the interpretability of the test results, will be illustrated through instructive case studies. We will also discuss some implementational aspects of the realisation of these methods in the TISEAN software package.<P><H3>Note:</H3><P>A version of this document has been published as a journal article:<blockquote> [<font color=red>Physica D <b>142</b> 346 (2000)</font>]</blockquote>which constitutes the correct referenceto this work.Preprint <a href="http://xxx.lanl.gov/abs/chao-dyn/9909037">paper versions</a> are also available.</P><P><P><BR> <HR><UL><A NAME="CHILD_LINKS"> </A><LI> <A NAME="tex2html31" HREF="node1.html#SECTION00010000000000000000">Introduction</A><LI> <A NAME="tex2html32" HREF="node2.html#SECTION00020000000000000000">Detecting weak nonlinearity</A><UL><LI> <A NAME="tex2html33" HREF="node3.html#SECTION00021000000000000000">Higher order statistics</A><LI> <A NAME="tex2html34" HREF="node4.html#SECTION00022000000000000000">Phase space observables</A></UL> <LI> <A NAME="tex2html35" HREF="node5.html#SECTION00030000000000000000">Surrogate data testing</A><UL><LI> <A NAME="tex2html36" HREF="node6.html#SECTION00031000000000000000">Typical vs. constrained realisations</A><LI> <A NAME="tex2html37" HREF="node7.html#SECTION00032000000000000000">The null hypothesis: model class vs. properties</A><LI> <A NAME="tex2html38" HREF="node8.html#SECTION00033000000000000000">Test design</A></UL> <LI> <A NAME="tex2html39" HREF="node9.html#SECTION00040000000000000000">Fourier based surrogates</A><UL><LI> <A NAME="tex2html40" HREF="node10.html#SECTION00041000000000000000">Rescaled Gaussian linear process</A><LI> <A NAME="tex2html41" HREF="node11.html#SECTION00042000000000000000">Flatness bias of AAFT surrogates</A><LI> <A NAME="tex2html42" HREF="node12.html#SECTION00043000000000000000">Iteratively refined surrogates</A><LI> <A NAME="tex2html43" HREF="node13.html#SECTION00044000000000000000">Example: Southern oscillation index</A><LI> <A NAME="tex2html44" HREF="node14.html#SECTION00045000000000000000">Periodicity artefacts</A><LI> <A NAME="tex2html45" HREF="node15.html#SECTION00046000000000000000">Iterative multivariate surrogates</A></UL> <LI> <A NAME="tex2html46" HREF="node16.html#SECTION00050000000000000000">General constrained randomisation</A><UL><LI> <A NAME="tex2html47" HREF="node17.html#SECTION00051000000000000000">Null hypotheses, constraints, and cost functions</A><LI> <A NAME="tex2html48" HREF="node18.html#SECTION00052000000000000000">Computational issues of simulated annealing</A><LI> <A NAME="tex2html49" HREF="node19.html#SECTION00053000000000000000">Example: avoiding periodicity artefacts</A><LI> <A NAME="tex2html50" HREF="node20.html#SECTION00054000000000000000">Combinatorial minimisation and accuracy</A><LI> <A NAME="tex2html51" HREF="node21.html#SECTION00055000000000000000">The curse of accuracy</A></UL> <LI> <A NAME="tex2html52" HREF="node22.html#SECTION00060000000000000000">Various Examples</A><UL><LI> <A NAME="tex2html53" HREF="node23.html#SECTION00061000000000000000">Including non-stationarity</A><LI> <A NAME="tex2html54" HREF="node24.html#SECTION00062000000000000000">Multivariate data</A><LI> <A NAME="tex2html55" HREF="node25.html#SECTION00063000000000000000">Uneven sampling</A><LI> <A NAME="tex2html56" HREF="node26.html#SECTION00064000000000000000">Spike trains</A></UL> <LI> <A NAME="tex2html57" HREF="node27.html#SECTION00070000000000000000">Questions of interpretation</A><UL><LI> <A NAME="tex2html58" HREF="node28.html#SECTION00071000000000000000">Non-dynamic nonlinearity</A><LI> <A NAME="tex2html59" HREF="node29.html#SECTION00072000000000000000">Non-stationarity</A></UL> <LI> <A NAME="tex2html60" HREF="node30.html#SECTION00080000000000000000">Conclusions: Testing a Hypothesis<BR> vs. Testing Against Surrogates</A><LI> <A NAME="tex2html61" HREF="node32.html#SECTION000100000000000000000">The TISEAN implementation</A><UL><LI> <A NAME="tex2html62" HREF="node33.html#SECTION000101000000000000000">Measures of nonlinearity</A><LI> <A NAME="tex2html63" HREF="node34.html#SECTION000102000000000000000">Iterative FFT surrogates</A><LI> <A NAME="tex2html64" HREF="node35.html#SECTION000103000000000000000">Annealed surrogates</A></UL> <LI> <A NAME="tex2html65" HREF="node36.html#SECTION000110000000000000000">References</A><LI> <A NAME="tex2html66" HREF="node37.html#SECTION000120000000000000000"> About this document ... </A></UL><HR><A NAME="tex2html29" HREF="node1.html"><IMG WIDTH=37 HEIGHT=24 ALIGN=BOTTOM ALT="next" SRC="next_motif.gif"></A> <IMG WIDTH=26 HEIGHT=24 ALIGN=BOTTOM ALT="up" SRC="up_motif_gr.gif"> <IMG WIDTH=63 HEIGHT=24 ALIGN=BOTTOM ALT="previous" SRC="previous_motif.gif"> <BR><B> Next:</B> <A NAME="tex2html30" HREF="node1.html">Introduction</A><P><ADDRESS><I>Thomas Schreiber <BR>Mon Aug 30 17:31:48 CEST 1999</I></ADDRESS></BODY></HTML>⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -