Changeset 608 for library/doc/html/tut_arx.html
- Timestamp:
- 09/12/09 11:41:43 (15 years ago)
- Files:
-
- 1 modified
Legend:
- Unmodified
- Added
- Removed
-
library/doc/html/tut_arx.html
r591 r608 68 68 <p>The <code>ARX</code> (AutoregRessive with eXogeneous input) model is defined as follows: </p> 69 69 <p class="formulaDsp"> 70 <img class="formulaDsp" alt="\[ y_t = \theta' \psi_t + \rho e_t \]" src="form_1 15.png"/>70 <img class="formulaDsp" alt="\[ y_t = \theta' \psi_t + \rho e_t \]" src="form_126.png"/> 71 71 </p> 72 <p> where <img class="formulaInl" alt="$y_t$" src="form_ 3.png"/> is the system output, <img class="formulaInl" alt="$[\theta,\rho]$" src="form_116.png"/> is vector of unknown parameters, <img class="formulaInl" alt="$\psi_t$" src="form_117.png"/> is an vector of data-dependent regressors, and noise <img class="formulaInl" alt="$e_t$" src="form_24.png"/> is assumed to be Normal distributed <img class="formulaInl" alt="$\mathcal{N}(0,1)$" src="form_118.png"/>.</p>72 <p> where <img class="formulaInl" alt="$y_t$" src="form_9.png"/> is the system output, <img class="formulaInl" alt="$[\theta,\rho]$" src="form_127.png"/> is vector of unknown parameters, <img class="formulaInl" alt="$\psi_t$" src="form_128.png"/> is an vector of data-dependent regressors, and noise <img class="formulaInl" alt="$e_t$" src="form_32.png"/> is assumed to be Normal distributed <img class="formulaInl" alt="$\mathcal{N}(0,1)$" src="form_129.png"/>.</p> 73 73 <p>Special cases include: </p> 74 74 <ul> … … 82 82 <dt>Information matrix </dt> 83 83 <dd>which is a sum of outer products <p class="formulaDsp"> 84 <img class="formulaDsp" alt="\[ V_t = \sum_{i=0}^{n} \left[\begin{array}{c}y_{t}\\ \psi_{t}\end{array}\right] \begin{array}{c} [y_{t}',\,\psi_{t}']\\ \\\end{array} \]" src="form_1 19.png"/>84 <img class="formulaDsp" alt="\[ V_t = \sum_{i=0}^{n} \left[\begin{array}{c}y_{t}\\ \psi_{t}\end{array}\right] \begin{array}{c} [y_{t}',\,\psi_{t}']\\ \\\end{array} \]" src="form_130.png"/> 85 85 </p> 86 86 </dd> 87 87 <dt>"Degree of freedom" </dt> 88 88 <dd>which is an accumulator of number of data records <p class="formulaDsp"> 89 <img class="formulaDsp" alt="\[ \nu_t = \sum_{i=0}^{n} 1 \]" src="form_1 20.png"/>89 <img class="formulaDsp" alt="\[ \nu_t = \sum_{i=0}^{n} 1 \]" src="form_131.png"/> 90 90 </p> 91 91 </dd> … … 94 94 On-line estimation</a></h2> 95 95 <p>For online estimation with stationary parameters can be easily achieved by collecting the sufficient statistics described above recursively.</p> 96 <p>Extension to non-stationaly parameters, <img class="formulaInl" alt="$ \theta_t , r_t $" src="form_1 21.png"/> can be achieved by operation called forgetting. This is an approximation of Bayesian filtering see [Kulhavy]. The resulting algorithm is defined by manipulation of sufficient statistics: </p>96 <p>Extension to non-stationaly parameters, <img class="formulaInl" alt="$ \theta_t , r_t $" src="form_132.png"/> can be achieved by operation called forgetting. This is an approximation of Bayesian filtering see [Kulhavy]. The resulting algorithm is defined by manipulation of sufficient statistics: </p> 97 97 <dl> 98 98 <dt>Information matrix </dt> 99 99 <dd>which is a sum of outer products <p class="formulaDsp"> 100 <img class="formulaDsp" alt="\[ V_t = \phi V_{t-1} + \left[\begin{array}{c}y_{t}\\ \psi_{t}\end{array}\right] \begin{array}{c} [y_{t}',\,\psi_{t}']\\ \\\end{array} +(1-\phi) V_0 \]" src="form_1 22.png"/>100 <img class="formulaDsp" alt="\[ V_t = \phi V_{t-1} + \left[\begin{array}{c}y_{t}\\ \psi_{t}\end{array}\right] \begin{array}{c} [y_{t}',\,\psi_{t}']\\ \\\end{array} +(1-\phi) V_0 \]" src="form_133.png"/> 101 101 </p> 102 102 </dd> 103 103 <dt>"Degree of freedom" </dt> 104 104 <dd>which is an accumulator of number of data records <p class="formulaDsp"> 105 <img class="formulaDsp" alt="\[ \nu_t = \phi \nu_{t-1} + 1 + (1-\phi) \nu_0 \]" src="form_1 23.png"/>105 <img class="formulaDsp" alt="\[ \nu_t = \phi \nu_{t-1} + 1 + (1-\phi) \nu_0 \]" src="form_134.png"/> 106 106 </p> 107 107 </dd> 108 108 </dl> 109 <p>where <img class="formulaInl" alt="$ \phi $" src="form_1 24.png"/> is the forgetting factor, typically <img class="formulaInl" alt="$ \phi \in [0,1]$" src="form_125.png"/> roughly corresponding to the effective length of the exponential window by relation:</p>109 <p>where <img class="formulaInl" alt="$ \phi $" src="form_135.png"/> is the forgetting factor, typically <img class="formulaInl" alt="$ \phi \in [0,1]$" src="form_136.png"/> roughly corresponding to the effective length of the exponential window by relation:</p> 110 110 <p class="formulaDsp"> 111 <img class="formulaDsp" alt="\[ \mathrm{win_length} = \frac{1}{1-\phi}\]" src="form_1 26.png"/>111 <img class="formulaDsp" alt="\[ \mathrm{win_length} = \frac{1}{1-\phi}\]" src="form_137.png"/> 112 112 </p> 113 <p> Hence, <img class="formulaInl" alt="$ \phi=0.9 $" src="form_1 27.png"/> corresponds to estimation on exponential window of effective length 10 samples.</p>114 <p>Statistics <img class="formulaInl" alt="$ V_0 , \nu_0 $" src="form_1 28.png"/> are called alternative statistics, their role is to stabilize estimation. It is easy to show that for zero data, the statistics <img class="formulaInl" alt="$ V_t , \nu_t $" src="form_129.png"/> converge to the alternative statistics.</p>113 <p> Hence, <img class="formulaInl" alt="$ \phi=0.9 $" src="form_138.png"/> corresponds to estimation on exponential window of effective length 10 samples.</p> 114 <p>Statistics <img class="formulaInl" alt="$ V_0 , \nu_0 $" src="form_139.png"/> are called alternative statistics, their role is to stabilize estimation. It is easy to show that for zero data, the statistics <img class="formulaInl" alt="$ V_t , \nu_t $" src="form_140.png"/> converge to the alternative statistics.</p> 115 115 <h2><a class="anchor" id="str"> 116 116 Structure estimation</a></h2> 117 <p>For this model, structure estimation is a form of model selection procedure. Specifically, we compare hypotheses that the data were generated by the full model with hypotheses that some regressors in vector <img class="formulaInl" alt="$\psi$" src="form_ 33.png"/> are redundant. The number of possible hypotheses is then the number of all possible combinations of all regressors.</p>118 <p>However, due to property known as nesting in exponential family, these hypotheses can be tested using only the posterior statistics. (This property does no hold for forgetting <img class="formulaInl" alt="$ \phi<1 $" src="form_1 30.png"/>). Hence, for low dimensional problems, this can be done by a tree search (method <a class="el" href="classbdm_1_1ARX.html#a16b02ae03316751664c22d59d90c1e34" title="Brute force structure estimation.">bdm::ARX::structure_est()</a>). Or more sophisticated algorithm [ref Ludvik]</p>117 <p>For this model, structure estimation is a form of model selection procedure. Specifically, we compare hypotheses that the data were generated by the full model with hypotheses that some regressors in vector <img class="formulaInl" alt="$\psi$" src="form_41.png"/> are redundant. The number of possible hypotheses is then the number of all possible combinations of all regressors.</p> 118 <p>However, due to property known as nesting in exponential family, these hypotheses can be tested using only the posterior statistics. (This property does no hold for forgetting <img class="formulaInl" alt="$ \phi<1 $" src="form_141.png"/>). Hence, for low dimensional problems, this can be done by a tree search (method <a class="el" href="classbdm_1_1ARX.html#a16b02ae03316751664c22d59d90c1e34" title="Brute force structure estimation.">bdm::ARX::structure_est()</a>). Or more sophisticated algorithm [ref Ludvik]</p> 119 119 <h2><a class="anchor" id="soft"> 120 120 Software Image</a></h2> … … 128 128 <h2><a class="anchor" id="try"> 129 129 How to try</a></h2> 130 <p>The best way to experiment with this object is to run matlab script <code>arx_test.m</code> located in directory <code></code>./library/tutorial. See <a class="el" href="arx_ui.html">Running experiment <code>estimator</code> with ARX data fields </a> for detailed description.</p>130 <p>The best way to experiment with this object is to run matlab script <code>arx_test.m</code> located in directory <code></code>./library/tutorial. See <a class="el" href="arx_ui.html">Running experiment <code>estimator</code> with ARX data fields this page is out of date, as the user info concept has been changed</a> for detailed description.</p> 131 131 <ul> 132 132 <li>In default setup, the parameters converge to the true values as expected. </li> … … 135 135 </ul> 136 136 </div> 137 <hr size="1"/><address style="text-align: right;"><small>Generated on Sun Aug 30 22:10:502009 for mixpp by 137 <hr size="1"/><address style="text-align: right;"><small>Generated on Tue Sep 8 22:11:32 2009 for mixpp by 138 138 <a href="http://www.doxygen.org/index.html"> 139 139 <img class="footer" src="doxygen.png" alt="doxygen"/></a> 1.6.1 </small></address>