📄 surrogates.html
字号:
<html> <head><title>Make srrogate data, also multivariate</title></head><body bgcolor="#ffffff"><h3>Make surrogate data, also multivariate</h3><font color=blue><tt>surrogates [-n# -i# -S -I# -o </tt><em>outfile</em><tt> -l#-x# -m# -c#[,#] -V# -h] </tt><em> file</em></font><blockquote> <br> <font color=blue><tt> -n </tt></font>number of surrogates (1) <br> <font color=blue><tt> -i </tt></font>number of iterations (until no change) <br> <font color=blue><tt> -S </tt></font>make spectrum exact rather than distribution <br> <font color=blue><tt> -I </tt></font>seed for random numbers <br> <font color=blue><tt> -o </tt></font><a href=../general.html#outfile>output file name</a>, just <font color=blue><tt> -o </tt></font>means <font color=blue><em>file</em></font><font color=blue><tt>_surr(_nnn)</tt></font> <br> <font color=blue><tt> -l </tt></font>number of points (whole file) <br> <font color=blue><tt> -x </tt></font>number of values to be skipped (0) <br> <font color=blue><tt> -m </tt></font>number of <a href=../general.html#columns>columns to be read</a> (1) <br> <font color=blue><tt> -c </tt></font><a href=../general.html#columns>columns to be read</a> (1) <br> <font color=blue><tt> -V </tt></font><a href=../general.html#verbosity>verbosity level</a> (0 = only fatal errors) <br> <font color=blue><tt> -h </tt></font>show usage message<p> <a href=../general.html#verbosity>verbosity level</a> (add what you want):<p> <font color=blue> 1</font> = input/output<br> <font color=blue> 2</font> = print # of iterations / discrepancy</blockquote>Reads data from a file and creates surrogate data with the same Fourieramplitudes and the same distribution of values. If more than one columns arespecified by either giving their number with <font color=blue><tt> -m </tt></font> or a commaseperated list with <font color=blue><tt> -c</tt></font>, multivariatesurrogates are prepared. In that case, also the relative Fourier phases of the channels are matched to those of the data.Since for finite length, distribution and spectral properties cannotbe guaranteed exactly at the same time, the default output containsthe iteration stage with the exact amplitudes. With the option<font color=blue><tt> -S </tt></font> set, the stage with the exact spectrum(resp. relative phases) is given. Asymptotically, the difference between both should converge to zero.<p>The setting <font color=blue><tt>-i0</tt></font> yields a random permutation,or, if the option<font color=blue><tt> -S </tt></font> set,an unrescaled FFT surrogate.The setting <font color=blue><tt>-i1</tt></font> yields a surrogate that is close to the result of the AAFT procedure, but not quite thesame. <p>Surrogates are written to <fontcolor=blue><tt>stdout</tt></font> by default. With the <fontcolor=blue><tt> -o </tt></font> option to <fontcolor=blue><em>file</em><tt>_surr_</tt><em>n</em></font>,n=1...<em>number</em>. For each surrogate, the iteration count and the rmsdiscrepancy between the exact spectrum and exact amplitude stages (normalisedto the rms amplitude of the data) are printed.<p><font color=red><b>Note:</b></font> The length of the sequence will betruncated to the largest sub-sequence factorizable with factors 2,3, and 5.This is necessary in order to use an efficient FFT.<p><font color=red><b>Note:</b></font> It is advisable to select a suitable sub-sequence to minimize end effects byusing <a href="endtoend.html">endtoend</a> before preparing surrogates. By nomeans should the data be zero-padded.<p><font color=red><b>Note:</b></font> Successive identical calls to this routinewill always yield the same surrogates, since "random" numbers on a computer aredeterministic. It is, however, possible to set the seed of the random numbergenerator explicitly by using the option <fontcolor=blue><tt> -I </tt></font>, so that calls with different seeds yielddifferent surogates.<p>See also the page on <a href="test.html">nonlinearity testing</a>.<p>This routine is based on<blockquote>T. Schreiber and A. Schmitz<br><a href="http://xxx.lanl.gov/abs/chao-dyn/9909041"><em>Improved surrogate data for nonlinearity tests</em></a><br><a href="http://ojps.aip.org/journals/doc/PRLTAO-home/top.html">Phys. Rev. Lett.</a> <b>77</b>, 635 (1996).</blockquote><p><hr><h3>Usage example</h3>Try also just running: <font color=blue><tt>gnuplot surrogates.gnu</tt></font> in the <font color=blue><tt>examples</tt></font> directory.<p><pre>gnuplot> <font color=blue><tt>plot 'spike.dat'</tt></font>gnuplot> <font color=blue><tt>plot '< surrogates spike.dat'</tt></font></pre><p><IMG SRC="spike.gif" alt="spike data"><p><IMG SRC="surrogates.gif" alt="spike surrogate"><p><a href="../contents.html">Table of Contents</a> * <a href="../../index.html" target="_top">TISEAN home</a></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 + -