ex3_answer1.html

非线性时间学列分析工具

<b><font color=red>Answers to time scales problems</font></b><br><br>

Notice that since <b>arch.dat</b> is non-stationary, no time averaged
information is meaningful! Once can compute the numbers, but they
cannot characterize the properties of the data!
<br>
<table noborder>
<tr> <td> </td><td> amplitude.dat</td><td>
whatisit.dat</td></tr> 
<tr><td>First zero of the auto-correlation function
</td><td>8-9 </td><td>10</td></tr>
<tr><td>exponential decay of envelope</td><td>data set too short</td><td> -0.0025</td></tr>
<tr><td>lag from 2-dim delay embedding</td><td> 8 </td><td>3</td></tr>
<tr><td> Theiler window </td><td> about 20 </td><td> about 20</td></tr>
</table>
<br>
The Theiler window should be half a period for such an almost periodic
pattern, but is suitablt chosen much larger, since typically not much
statistics is lost, as long as it remains smaller than about 10% of
the data set size.
<br><br>
Usage of <a href="../docs_f/stp.html"> stp </a>, e.g.,: <br>
<font color=green>  plot '&#60; stp whatisit.dat -d5 -m3 -t50' </font>
<br><br>