📄 node15.html

📁 an analysis software with souce code for the time series with methods based on the theory of nonline
💻 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>Iterative multivariate surrogates</TITLE>
<META NAME="description" CONTENT="Iterative multivariate surrogates">
<META NAME="keywords" CONTENT="Surrogates">
<META NAME="resource-type" CONTENT="document">
<META NAME="distribution" CONTENT="global">
<LINK REL=STYLESHEET HREF="Surrogates.css" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/Surrogates.css">
</HEAD>
<BODY bgcolor=#ffffff LANG="EN" >
 <A NAME="tex2html220" HREF="node16.html" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node16.html"><IMG WIDTH=37 HEIGHT=24 ALIGN=BOTTOM ALT="next" SRC="next_motif.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/next_motif.gif"></A> <A NAME="tex2html218" HREF="node9.html" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node9.html"><IMG WIDTH=26 HEIGHT=24 ALIGN=BOTTOM ALT="up" SRC="up_motif.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/up_motif.gif"></A> <A NAME="tex2html214" HREF="node14.html" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node14.html"><IMG WIDTH=63 HEIGHT=24 ALIGN=BOTTOM ALT="previous" SRC="previous_motif.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/previous_motif.gif"></A>   <BR>
<B> Next:</B> <A NAME="tex2html221" HREF="node16.html" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node16.html">General constrained randomisation</A>
<B>Up:</B> <A NAME="tex2html219" HREF="node9.html" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node9.html">Fourier based surrogates</A>
<B> Previous:</B> <A NAME="tex2html215" HREF="node14.html" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node14.html">Periodicity artefacts</A>
<BR> <P>
<H2><A NAME="SECTION00046000000000000000">Iterative multivariate surrogates</A></H2>
<A NAME="secmulti1">&#160;</A>
<P>
<P>
A natural generalisation of the null hypothesis of a Gaussian linear stochastic
process is that of a multivariate process of the same kind. In this case, the
process is determined by giving the cross-spectrum in
addition to the power spectrum of each of the channels. In Ref.&nbsp;[<A HREF="node36.html#multi" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node36.html#multi">34</A>],
it has been pointed out that phase randomised surrogates are readily produced
by multiplying the Fourier phases of each of the channels by the same set of 
random phases since the cross-spectrum reflects relative phases only.
The authors of Ref.&nbsp;[<A HREF="node36.html#multi" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node36.html#multi">34</A>] did not discuss the possibility to combine 
multivariate phase randomisation with an amplitude adjustment step.
The extension of the iterative refinement scheme introduced in
Sec.&nbsp;<A HREF="node12.html#seciterative" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node12.html#seciterative">4.3</A> to the multivariate case is relatively
straightforward. Since deviations from a Gaussian distribution are very common
and may occur due to a simple invertible rescaling due to the measurement
process, we want to give the algorithm here.
<P>
Recall that the iterative scheme consists of two procedures which are applied
in an alternating fashion until convergence to a fixed point is achieved. The
amplitude adjustment procedure by rank ordering (<A HREF="node12.html#eqstep2" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node12.html#eqstep2">13</A>) is readily
applied to each channel individually. However, the spectral adjustment in the
Fourier domain has to be modified. Let us introduce a second index in order to
denote the <I>M</I> different channels of a multivariate time series <IMG WIDTH=274 HEIGHT=24 ALIGN=MIDDLE ALT="tex2html_wrap_inline2084" SRC="img71.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/img71.gif">. The change that has to be applied
to the ``filter'' step, Eq.(<A HREF="node12.html#eqstep1" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node12.html#eqstep1">12</A>), is that the phases
<IMG WIDTH=32 HEIGHT=22 ALIGN=MIDDLE ALT="tex2html_wrap_inline2086" SRC="img72.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/img72.gif"> have to be replaced by phases <IMG WIDTH=30 HEIGHT=22 ALIGN=MIDDLE ALT="tex2html_wrap_inline2088" SRC="img73.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/img73.gif"> with the
following properties. (We have dropped the superscript (<I>i</I>) for convenience.)
The replacement should be minimal in the least
squares sense, that is, it should minimise
<BR><A NAME="eqsum">&#160;</A><IMG WIDTH=500 HEIGHT=46 ALIGN=BOTTOM ALT="equation1058" SRC="img74.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/img74.gif"><BR>
Also, the new phases must implement the same phase differences exhibited by the
corresponding phases <IMG WIDTH=141 HEIGHT=27 ALIGN=MIDDLE ALT="tex2html_wrap_inline2092" SRC="img75.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/img75.gif"> of the data:
<BR><IMG WIDTH=500 HEIGHT=20 ALIGN=BOTTOM ALT="equation1060" SRC="img76.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/img76.gif"><BR>
The last equation can be fulfilled by setting <IMG WIDTH=119 HEIGHT=22 ALIGN=MIDDLE ALT="tex2html_wrap_inline2094" SRC="img77.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/img77.gif">. With this, we have <IMG WIDTH=276 HEIGHT=33 ALIGN=MIDDLE ALT="tex2html_wrap_inline2096" SRC="img78.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/img78.gif"> which is extremal when
<BR><IMG WIDTH=500 HEIGHT=46 ALIGN=BOTTOM ALT="equation1062" SRC="img79.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/img79.gif"><BR>
The minimum is selected by taking <IMG WIDTH=16 HEIGHT=14 ALIGN=MIDDLE ALT="tex2html_wrap_inline2098" SRC="img80.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/img80.gif"> in the correct quadrant.
<P>
<P><blockquote><A NAME="944">&#160;</A><IMG WIDTH=345 HEIGHT=388 ALIGN=BOTTOM ALT="figure1056" SRC="img69.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/img69.gif"><BR>
<STRONG>Figure:</STRONG> <A NAME="figbdat">&#160;</A>
   Simultaneous surrogates for a bi-variate time series. The upper two panels
   show simultaneous recordings of the breath rate and the instantaneous heart
   rate of a human. The lower two panels show surrogate sequences that preserve
   the individual distributions and power spectra as well as the
   cross-correlation function between heart and breath rate. The most prominent
   difference between data and surrogates is the lack of coherence in the
   surrogate breath rate.  <BR>
</blockquote><P>
<P><blockquote><A NAME="946">&#160;</A><IMG WIDTH=364 HEIGHT=287 ALIGN=BOTTOM ALT="figure1057" SRC="img70.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/img70.gif"><BR>
<STRONG>Figure:</STRONG> <A NAME="figbx">&#160;</A>
   Cross-correlation functions for the bi-variate data shown in
   Fig.&nbsp;<A HREF="node15.html#figbdat" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node15.html#figbdat">8</A> (upper panel), and a surrogate that preserves the
   individual spectra and distributions as well as the relative Fourier phases
   (middle). The lower panel shows the same for surrogates prepared for each
   channel individually, that is, without explicitly preserving the
   cross-correlation structure.<BR>
</blockquote><P>

As an example, let us generate a surrogate sequence for a simultaneous
recording of the breath rate and the instantaneous heart rate of a
human during sleep. The data is again taken from data set B of the
Santa Fe Institute time series contest&nbsp;[<A HREF="node36.html#gold" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node36.html#gold">31</A>]. The 1944 data
points are an end-point matched sub-sequence of the data used as a
multivariate example in Ref.&nbsp;[<A HREF="node36.html#anneal" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node36.html#anneal">26</A>]. In the latter study, which
will be commented on in Sec.&nbsp;<A HREF="node24.html#secmulti2" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node24.html#secmulti2">6.2</A> below, the breath rate
signal had been considered to be an input and therefore not been
randomised. Here, we will randomise both channels under the condition
that their individual spectra as well as their cross-correlation
function are preserved as well as possible while matching the
individual distributions exactly. The iterative scheme introduced
above took 188 iterations to converge to a fixed point.  The data and
a bi-variate surrogate is shown in Fig.&nbsp;<A HREF="node15.html#figbdat" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node15.html#figbdat">8</A>.  In
Fig.&nbsp;<A HREF="node15.html#figbx" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node15.html#figbx">9</A>, the cross-correlation functions of the data and one
surrogate are plotted. Also, for comparison, the same for two individual
surrogates of the two channels. The most striking
difference between data and surrogates is that the coherence of the
breath rate is lost. Thus, it is indeed reasonable to exclude the
nonlinear structure in the breath dynamics from a further analysis of
the heart rate by taking the breath rate as a given input signal.
Such an analysis is however beyond the scope of the method discussed
in this section. First of all, specifying the full cross-correlation
function to a fixed signal plus the autocorrelation function
over-specifies the problem and there is no room for randomisation. In
Sec.&nbsp;<A HREF="node24.html#secmulti2" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node24.html#secmulti2">6.2</A> below, we will therefore revisit this
problem. With the general constrained randomisation scheme to be
introduced below, it will be possible to specify a limited number of
lags of the auto- and cross-correlation functions.
<P>
<HR><A NAME="tex2html220" HREF="node16.html" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node16.html"><IMG WIDTH=37 HEIGHT=24 ALIGN=BOTTOM ALT="next" SRC="next_motif.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/next_motif.gif"></A> <A NAME="tex2html218" HREF="node9.html" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node9.html"><IMG WIDTH=26 HEIGHT=24 ALIGN=BOTTOM ALT="up" SRC="up_motif.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/up_motif.gif"></A> <A NAME="tex2html214" HREF="node14.html" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node14.html"><IMG WIDTH=63 HEIGHT=24 ALIGN=BOTTOM ALT="previous" SRC="previous_motif.gif" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/previous_motif.gif"></A>   <BR>
<B> Next:</B> <A NAME="tex2html221" HREF="node16.html" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node16.html">General constrained randomisation</A>
<B>Up:</B> <A NAME="tex2html219" HREF="node9.html" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node9.html">Fourier based surrogates</A>
<B> Previous:</B> <A NAME="tex2html215" HREF="node14.html" tppabs="http://www.mpipks-dresden.mpg.de/~tisean/TISEAN_2.0/docs/surropaper/node14.html">Periodicity artefacts</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 + -