1
<?xml version="1.0" encoding="UTF-8"?>
2
<refentry xmlns="http://docbook.org/ns/docbook" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:svg="http://www.w3.org/2000/svg" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:db="http://docbook.org/ns/docbook" version="5.0-subset Scilab" xml:lang="ja" xml:id="phc">
5
<refpurpose> Markovian表現</refpurpose>
9
<synopsis>[H,F,G]=phc(hk,d,r)</synopsis>
37
<para>Markovianモデルの行列</para>
45
確率過程の共分散系列から構築されたハンケル行列から
47
Markovian表現の行列<literal>H, F, G</literal>を計算する関数.
52
<programlisting role="example"><![CDATA[
53
//This example may usefully be compared with the results from
54
//the 'levin' macro (see the corresponding help and example)
56
//We consider the process defined by two sinusoids (1Hz and 2 Hz)
57
//in additive Gaussian noise (this is the observation);
58
//the simulated process is sampled at 10 Hz.
60
t=0:.1:100;rand('normal');
61
y=sin(2*%pi*t)+sin(2*%pi*2*t);y=y+rand(y);plot(t,y)
68
//hankel matrix from the covariance sequence
69
//(we can choose to take more information from covariance
70
//by taking greater n and m; try it to compare the results !
75
//compute the Markov representation (mh,mf,mg)
76
//We just take here a state dimension equal to 4 :
77
//this is the rather difficult problem of estimating the order !
79
//(the observation dimension is here equal to one)
82
[mh,mf,mg]=phc(h,1,ns);
84
//verify that the spectrum of mf contains the
85
//frequency spectrum of the observed process y
86
//(remember that y is sampled -in our example
87
//at 10Hz (T=0.1s) so that we need
88
//to retrieve the original frequencies through the log
89
//and correct scaling by the frequency sampling)
94
//now we get the estimated spectrum
98
<refsection role="see also">
100
<simplelist type="inline">
102
<link linkend="levin">levin</link>