root/doc/html/arx.html @ 270

Revision 270, 3.2 kB (checked in by smidl, 15 years ago)

Changes in the very root classes!
* rv and rvc are no longer compulsory,
* samplecond does not return ll
* BM has drv

Line 
1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
2<html><head><meta http-equiv="Content-Type" content="text/html;charset=UTF-8">
3<title>mixpp: Example of ARX model estimation</title>
4<link href="doxygen.css" rel="stylesheet" type="text/css">
5<link href="tabs.css" rel="stylesheet" type="text/css">
6</head><body>
7<!-- Generated by Doxygen 1.5.6 -->
8<div class="navigation" id="top">
9  <div class="tabs">
10    <ul>
11      <li><a href="index.html"><span>Main&nbsp;Page</span></a></li>
12      <li><a href="pages.html"><span>Related&nbsp;Pages</span></a></li>
13      <li><a href="modules.html"><span>Modules</span></a></li>
14      <li><a href="namespaces.html"><span>Namespaces</span></a></li>
15      <li><a href="classes.html"><span>Classes</span></a></li>
16      <li><a href="files.html"><span>Files</span></a></li>
17    </ul>
18  </div>
19</div>
20<div class="contents">
21<h1><a class="anchor" name="arx">Example of ARX model estimation </a></h1>Here, we use the <code>ARX</code> class to estimate parameters and structure. ARX model is defined as follows: <p class="formulaDsp">
22<img class="formulaDsp" alt="\[ y_t = \theta' \psi_t + \rho e_t \]" src="form_101.png">
23<p>
24 where <img class="formulaInl" alt="$y_t$" src="form_94.png"> is the system output, <img class="formulaInl" alt="$[\theta,\rho]$" src="form_95.png"> is vector of unknown parameters, <img class="formulaInl" alt="$\psi_t$" src="form_102.png"> is an vector of data-dependent regressors, and noise <img class="formulaInl" alt="$e_t$" src="form_54.png"> is assumed to be Normal distributed <img class="formulaInl" alt="$\mathcal{N}(0,1)$" src="form_97.png">.<p>
25Special cases include:...<h2><a class="anchor" name="math">
26Mathematical background:</a></h2>
27This particular model belongs to the exponential family, hence it has conjugate distribution of the Gauss-inverse-Wishart form (class egiw). See, [reference] for details.<p>
28For this model, structure estimation is a form of model selection procedure. Specifically, we compare hypotheses that the data were generated by the full model with hypotheses that some regressors in vector <img class="formulaInl" alt="$\psi$" src="form_67.png"> are redundant. The number of possible hypotheses is then the number of all possible combinations of all regressors.<h2><a class="anchor" name="soft">
29Software implementation:</a></h2>
30Estimation with this class of model is perfromed by class ARX which is derived from class BMEF (estimation of exponential family). The posterior density ( ARX::_epdf() ) is class egiw, which represents Gauss-inverse-Wishart density.<p>
31Structure estimation is implemented in method ARX::structure_est() which uses brute force tree search approach.<h2><a class="anchor" name="exa">
32Examples of Use:</a></h2>
33There are many ways how to use the object.<ul>
34<li>Pure C++, as it is used in unit testing of the class arx, <a class="el" href="arx__test_8cpp.html">arx_test.cpp</a></li><li>C++ application with UI configuration file, arx_test_ui</li><li>Matlab interface, arx_matlab </li></ul>
35</div>
36<hr size="1"><address style="text-align: right;"><small>Generated on Wed Feb 11 23:33:57 2009 for mixpp by&nbsp;
37<a href="http://www.doxygen.org/index.html">
38<img src="doxygen.png" alt="doxygen" align="middle" border="0"></a> 1.5.6 </small></address>
39</body>
40</html>
Note: See TracBrowser for help on using the browser.