← Back to branch summary

~ubuntu-branches/ubuntu/hoary/scilab/hoary

« back to all changes in this revision

Viewing changes to man/eng/arma/armax1.xml

  • Committer: Bazaar Package Importer
  • Author(s): Torsten Werner
  • Date: 2005-01-09 22:58:21 UTC
  • mfrom: (1.1.1 upstream)
  • Revision ID: james.westby@ubuntu.com-20050109225821-473xr8vhgugxxx5j
Tags: 3.0-12
http://bugs.debian.org/255869
changed configure.in to build scilab's own malloc.o, closes: #255869

Show diffs side-by-side

added added

removed removed

Lines of Context:
man/eng/arma/armax1.xml
 
1
<?xml version="1.0" encoding="ISO-8859-1" standalone="no"?>
 
2
<!DOCTYPE MAN SYSTEM "../../manrev.dtd">
 
3
<MAN>
 
4
  <LANGUAGE>eng</LANGUAGE>
 
5
  <TITLE>armax1</TITLE>
 
6
  <TYPE>Scilab Function</TYPE>
 
7
  <DATE>April 1993</DATE>
 
8
  <SHORT_DESCRIPTION name="armax1"> armax identification</SHORT_DESCRIPTION>
 
9
  <CALLING_SEQUENCE>
 
10
    <CALLING_SEQUENCE_ITEM>[arc,resid]=armax1(r,s,q,y,u [,b0f])  </CALLING_SEQUENCE_ITEM>
 
11
  </CALLING_SEQUENCE>
 
12
  <PARAM>
 
13
    <PARAM_INDENT>
 
14
      <PARAM_ITEM>
 
15
        <PARAM_NAME>y</PARAM_NAME>
 
16
        <PARAM_DESCRIPTION>
 
17
          <SP>: output signal</SP>
 
18
        </PARAM_DESCRIPTION>
 
19
      </PARAM_ITEM>
 
20
      <PARAM_ITEM>
 
21
        <PARAM_NAME>u</PARAM_NAME>
 
22
        <PARAM_DESCRIPTION>
 
23
          <SP>: input signal</SP>
 
24
        </PARAM_DESCRIPTION>
 
25
      </PARAM_ITEM>
 
26
      <PARAM_ITEM>
 
27
        <PARAM_NAME>r,s,q</PARAM_NAME>
 
28
        <PARAM_DESCRIPTION>
 
29
          <SP>: auto regression orders with r &gt;=0, s &gt;=-1.</SP>
 
30
        </PARAM_DESCRIPTION>
 
31
      </PARAM_ITEM>
 
32
      <PARAM_ITEM>
 
33
        <PARAM_NAME>b0f</PARAM_NAME>
 
34
        <PARAM_DESCRIPTION>
 
35
          <SP>: optional parameter. Its default value is 0 and it means that the coefficient b0 must be identified. if bof=1 the b0 is supposed to be zero and is not identified</SP>
 
36
        </PARAM_DESCRIPTION>
 
37
      </PARAM_ITEM>
 
38
      <PARAM_ITEM>
 
39
        <PARAM_NAME>arc</PARAM_NAME>
 
40
        <PARAM_DESCRIPTION>
 
41
          <SP>: is tlist with type &quot;ar&quot; and fields a,  b,  d,  ny,  nu,  sig</SP>
 
42
          <PARAM_INDENT>
 
43
            <PARAM_ITEM>
 
44
              <PARAM_NAME>a</PARAM_NAME>
 
45
              <PARAM_DESCRIPTION>
 
46
                <SP>: is the vector [1,a1,...,a_r]</SP>
 
47
              </PARAM_DESCRIPTION>
 
48
            </PARAM_ITEM>
 
49
            <PARAM_ITEM>
 
50
              <PARAM_NAME>b</PARAM_NAME>
 
51
              <PARAM_DESCRIPTION>
 
52
                <SP>: is the vector [b0,......,b_s]</SP>
 
53
              </PARAM_DESCRIPTION>
 
54
            </PARAM_ITEM>
 
55
            <PARAM_ITEM>
 
56
              <PARAM_NAME>d</PARAM_NAME>
 
57
              <PARAM_DESCRIPTION>
 
58
                <SP>: is the vector [1,d1,....,d_q]</SP>
 
59
              </PARAM_DESCRIPTION>
 
60
            </PARAM_ITEM>
 
61
            <PARAM_ITEM>
 
62
              <PARAM_NAME>sig</PARAM_NAME>
 
63
              <PARAM_DESCRIPTION>
 
64
                <SP>:   resid=[ sig*echap(1),....,];</SP>
 
65
              </PARAM_DESCRIPTION>
 
66
            </PARAM_ITEM>
 
67
          </PARAM_INDENT>
 
68
        </PARAM_DESCRIPTION>
 
69
      </PARAM_ITEM>
 
70
    </PARAM_INDENT>
 
71
  </PARAM>
 
72
  <DESCRIPTION>
 
73
    <P>
 
74
    armax1 is used to identify the coefficients of a 1-dimensional 
 
75
    ARX process:</P>
 
76
    <VERBATIM>
 
77
<![CDATA[
 
78
   A(z^-1)y= B(z^-1)u + D(z^-1)sig*e(t)
 
79
   e(t) is a 1-dimensional white noise with variance 1.
 
80
   A(z)= 1+a1*z+...+a_r*z^r; ( r=0 => A(z)=1)
 
81
   B(z)= b0+b1*z+...+b_s z^s ( s=-1 => B(z)=0)
 
82
   D(z)= 1+d1*z+...+d_q*z^q  ( q=0 => D(z)=1)
 
83
   ]]>
 
84
    </VERBATIM>
 
85
    <P>
 
86
    for the method, see Eykhoff in trends and progress in system identification) page 96.
 
87
    with 
 
88
            z(t)=[y(t-1),..,y(t-r),u(t),...,
 
89
                  u(t-s),e(t-1),...,e(t-q)] 
 
90
    and
 
91
            coef= [-a1,..,-ar,b0,...,b_s,d1,...,d_q]'
 
92
            y(t)= coef'* z(t) + sig*e(t).</P>
 
93
    <P>
 
94
    a sequential version of the AR estimation where e(t-i) is replaced 
 
95
    by an estimated value is used (RLLS). With q=0 this method is exactly
 
96
    a sequential version of armax</P>
 
97
  </DESCRIPTION>
 
98
  <AUTHOR>J.-Ph.C; ;   </AUTHOR>
 
99
</MAN>