<?xml version="1.0" encoding="ISO-8859-1"?><article xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<front>
<journal-meta>
<journal-id>1409-2433</journal-id>
<journal-title><![CDATA[Revista de Matemática Teoría y Aplicaciones]]></journal-title>
<abbrev-journal-title><![CDATA[Rev. Mat]]></abbrev-journal-title>
<issn>1409-2433</issn>
<publisher>
<publisher-name><![CDATA[Centro de Investigaciones en Matemática Pura y Aplicada (CIMPA) y Escuela de Matemática, San José, Costa Rica.]]></publisher-name>
</publisher>
</journal-meta>
<article-meta>
<article-id>S1409-24332014000100006</article-id>
<title-group>
<article-title xml:lang="es"><![CDATA[Gnm-nipals: estimación general no métrica y no lineal por mínimos cuadrados parciales iterativos]]></article-title>
<article-title xml:lang="en"><![CDATA[Gnm-nipals: general nonmetric-nonlinear estimation by iterative partial least squares]]></article-title>
</title-group>
<contrib-group>
<contrib contrib-type="author">
<name>
<surname><![CDATA[Aluja]]></surname>
<given-names><![CDATA[Tomás]]></given-names>
</name>
<xref ref-type="aff" rid="A01"/>
</contrib>
<contrib contrib-type="author">
<name>
<surname><![CDATA[González]]></surname>
<given-names><![CDATA[Víctor Manuel]]></given-names>
</name>
<xref ref-type="aff" rid="A02"/>
</contrib>
</contrib-group>
<aff id="A01">
<institution><![CDATA[,Universitat Politècnica de Catalunya  ]]></institution>
<addr-line><![CDATA[ Barcelona]]></addr-line>
<country>España</country>
</aff>
<aff id="A02">
<institution><![CDATA[,Universitat Politècnica de Catalunya  ]]></institution>
<addr-line><![CDATA[ Barcelona]]></addr-line>
<country>España</country>
</aff>
<pub-date pub-type="pub">
<day>00</day>
<month>06</month>
<year>2014</year>
</pub-date>
<pub-date pub-type="epub">
<day>00</day>
<month>06</month>
<year>2014</year>
</pub-date>
<volume>21</volume>
<numero>1</numero>
<fpage>85</fpage>
<lpage>106</lpage>
<copyright-statement/>
<copyright-year/>
<self-uri xlink:href="http://www.scielo.sa.cr/scielo.php?script=sci_arttext&amp;pid=S1409-24332014000100006&amp;lng=en&amp;nrm=iso"></self-uri><self-uri xlink:href="http://www.scielo.sa.cr/scielo.php?script=sci_abstract&amp;pid=S1409-24332014000100006&amp;lng=en&amp;nrm=iso"></self-uri><self-uri xlink:href="http://www.scielo.sa.cr/scielo.php?script=sci_pdf&amp;pid=S1409-24332014000100006&amp;lng=en&amp;nrm=iso"></self-uri><abstract abstract-type="short" xml:lang="es"><p><![CDATA[En este trabajo se desarrolla GNM-NIPALS para formar parte de los métodos NM-PLS, el cual permite cuantificar las variables cualitativas de una matriz de datos mixtos mediante una función lineal de k componentes principales, tipo reconstitución, maximizando la inercia en el plano kdimensional asociado al ACP de la matriz así cuantificada. Es entonces una generalización del algoritmo NM-NIPALS que usa solo la primera componente principal en la cuantificación de variables cualitativas. De la maximización y positividad de la razón de correlación entre cada variable cualitativa y la función reconstituida, se tiene que la inercia acumulada en el plano k-dimensional asociado a la función de cuantificación del mismo rango, es mayor o igual que la generada en planos de igual dimensión pero con funciones de cuantificación de diferente rango. Con las k componentes principales asociadas a la matriz así cuantificada, se desarrolla el análisis de inercia saturada para evaluar si aún existe una dimensión k* < k, a partir de la cual la inercia acumulada en los ejes de orden igual o superior ya esta explicada, caso en el cual la función de cuantificación definitiva es de rango menor (k*).]]></p></abstract>
<abstract abstract-type="short" xml:lang="en"><p><![CDATA[This paper develops GNM-NIPALS as an extension of the NM-PLS methods, which allows to quantify the qualitative variables of mixed data, by means of the reconstitution function using the first k principal components, maximizing the inertia in the plane k subspace associated with the PCA of the quantified matrix. It generalizes the NM-NIPALS algorithm in the sense that the latter only uses the first principal component in the quantification of qualitative variables. From the maximization and positivity of the correlation ratio between each qualitative variable and the reconstituted function, we have that the accumulated inertia on the kdimensional subspace associated to the quantification function of the same range is greater than or equal to the one generated on subspaces of equal dimension, but with quantification functions of different range. With the k principal components associated to the quantified matrix, a saturated inertia analysis is performed to evaluate if a dimension k* < k still exists, from which the accumulated inertia on the axes of equal or superior order is already explained, in which case the definitive quantification function is of lesser range (k*).]]></p></abstract>
<kwd-group>
<kwd lng="es"><![CDATA[NM-PLS]]></kwd>
<kwd lng="es"><![CDATA[ACP]]></kwd>
<kwd lng="es"><![CDATA[datos mixtos]]></kwd>
<kwd lng="es"><![CDATA[cuantificación]]></kwd>
<kwd lng="es"><![CDATA[k-dimensional]]></kwd>
<kwd lng="es"><![CDATA[inercia saturada]]></kwd>
<kwd lng="es"><![CDATA[maximal]]></kwd>
<kwd lng="es"><![CDATA[razón correlación]]></kwd>
<kwd lng="en"><![CDATA[NM-PLS]]></kwd>
<kwd lng="en"><![CDATA[PCA]]></kwd>
<kwd lng="en"><![CDATA[mixed data]]></kwd>
<kwd lng="en"><![CDATA[quantification]]></kwd>
<kwd lng="en"><![CDATA[k-dimensional]]></kwd>
<kwd lng="en"><![CDATA[saturated inertia]]></kwd>
<kwd lng="en"><![CDATA[maximal]]></kwd>
<kwd lng="en"><![CDATA[correlation ratio]]></kwd>
</kwd-group>
</article-meta>
</front><body><![CDATA[ <div style="text-align: justify;">     <div style="text-align: center;"><font  style="font-family: Verdana; font-weight: bold;" size="4"> Gnm-nipals: estimaci&oacute;n general no m&eacute;trica y no lineal por m&iacute;nimos cuadrados parciales iterativos</font>    <br> </div> <font style="font-family: Verdana;" size="2"></font>    <br>     <div style="text-align: center;"><font  style="font-family: Verdana; font-weight: bold;" size="4">Gnm-nipals: general nonmetric-nonlinear estimation by iterative partial least squares</font><font style="font-family: Verdana; font-weight: bold;"  size="3"> </font>    <br> </div> <font style="font-family: Verdana;" size="2"></font>    <br>     <div style="text-align: center;"><font style="font-family: Verdana;"  size="2">Tom&aacute;s Aluja<sup><a href="#1">*</a><a name="3"></a>*</sup></font><font  style="font-family: Verdana;" size="2"> V&iacute;ctor Manuel Gonz&aacute;lez<sup><a href="#2">&#8224;</a><a name="4"></a>*</sup></font>    <br> </div> <font style="font-family: Verdana;" size="2"></font>    <br> <small><span style="font-family: Verdana;"><a name="Correspondencia2"></a>*<a  href="#Correspondencia1">Direcci&oacute;n para correspondencia</a></span></small><a  href="#Correspondencia1">:</a>    ]]></body>
<body><![CDATA[<br> <hr style="width: 100%; height: 2px;"><font  style="font-family: Verdana;" size="2"></font><font  style="font-family: Verdana; font-weight: bold;" size="3">Resumen</font>    <br> <font style="font-family: Verdana;" size="2"></font>    <br> <font style="font-family: Verdana;" size="2">En este trabajo se desarrolla GNM-NIPALS para formar parte de los m&eacute;todos NM-PLS, el cual permite cuantificar las variables cualitativas de una matriz de datos mixtos mediante una funci&oacute;n lineal de k componentes principales, tipo reconstituci&oacute;n, maximizando la inercia en el plano kdimensional asociado al ACP de la matriz as&iacute; cuantificada. Es entonces una generalizaci&oacute;n del algoritmo NM-NIPALS que usa solo la primera componente principal en la cuantificaci&oacute;n de variables cualitativas. De la maximizaci&oacute;n y positividad de la raz&oacute;n de correlaci&oacute;n entre cada variable cualitativa y la funci&oacute;n reconstituida, se tiene que la inercia acumulada en el plano k-dimensional asociado a la funci&oacute;n de cuantificaci&oacute;n del mismo rango, es mayor o igual que la generada en planos de igual dimensi&oacute;n pero con funciones de cuantificaci&oacute;n de diferente rango. Con las k componentes principales asociadas a la matriz as&iacute; cuantificada, se desarrolla el an&aacute;lisis de inercia saturada para evaluar si a&uacute;n existe una dimensi&oacute;n k* &lt; k, a partir de la cual la inercia acumulada en los ejes de orden igual o superior ya esta explicada, caso en el cual la funci&oacute;n de cuantificaci&oacute;n definitiva es de rango menor (k*).</font>    <br> <font style="font-family: Verdana;" size="2"></font>    <br> <font style="font-family: Verdana;" size="2"><span  style="font-weight: bold;">Palabras clave</span>: NM-PLS, ACP, datos mixtos, cuantificaci&oacute;n, k-dimensional, inercia saturada, maximal, raz&oacute;n correlaci&oacute;n.</font>    <br> <font style="font-family: Verdana;" size="2"></font>    <br> <font style="font-family: Verdana; font-weight: bold;" size="3">Abstract</font>    <br> <font style="font-family: Verdana;" size="2"></font>    <br> <font style="font-family: Verdana;" size="2">This paper develops GNM-NIPALS as an extension of the NM-PLS methods, which allows to quantify the qualitative variables of mixed data, by means of the reconstitution function using the first k principal components, maximizing the inertia in the plane k subspace associated with the PCA of the quantified matrix. It generalizes the NM-NIPALS algorithm in the sense that the latter only uses the first principal component in the quantification of qualitative variables. From the maximization and positivity of the correlation ratio between each qualitative variable and the reconstituted function, we have that the accumulated inertia on the kdimensional subspace associated to the quantification function of the same range is greater than or equal to the one generated on subspaces of equal dimension, but with quantification functions of different range. With the k principal components associated to the quantified matrix, a saturated inertia analysis is performed to evaluate if a dimension k* &lt; k still exists, from which the accumulated inertia on the axes of equal or superior order is already explained, in which case the definitive quantification function is of lesser range (k*).</font>    <br> <font style="font-family: Verdana;" size="2"></font>    ]]></body>
<body><![CDATA[<br> <font style="font-family: Verdana;" size="2"><span  style="font-weight: bold;">Keywords</span>: NM-PLS, PCA, mixed data, quantification, k-dimensional, saturated inertia, maximal, correlation ratio.</font>    <br> <font style="font-family: Verdana;" size="2"></font>    <br> <font style="font-family: Verdana;" size="2"><span  style="font-weight: bold;">Mathematics Subject Classification</span>: 62H25.</font>    <br> <hr style="width: 100%; height: 2px;"><font  style="font-family: Verdana;" size="2"></font><font  style="font-family: Verdana; font-weight: bold;" size="3">    <br> </font><small><font style="font-family: Verdana;" size="3"><small>Ver contenido disponible en pdf</small></font></small><font  style="font-family: Verdana; font-weight: bold;" size="3">    <br>     <br> </font> <hr style="width: 100%; height: 2px;"><font  style="font-family: Verdana; font-weight: bold;" size="3">Referencias</font>    <br> <font style="font-family: Verdana;" size="2"></font>    <!-- ref --><br> <font style="font-family: Verdana;" size="2">[1] Aluja, T.;Morineau, A. (1999) <span style="font-style: italic;">Aprender de los Datos: El An&aacute;lisis de Componentes Principales.</span> EUB S.L, Barcelona.    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[&#160;<a href="javascript:void(0);" onclick="javascript: window.open('/scielo.php?script=sci_nlinks&ref=1943802&pid=S1409-2433201400010000600001&lng=','','width=640,height=500,resizable=yes,scrollbars=1,menubar=yes,');">Links</a>&#160;]<!-- end-ref --></font>    ]]></body>
<body><![CDATA[<br> <font style="font-family: Verdana;" size="2"></font>    <!-- ref --><br> <font style="font-family: Verdana;" size="2">[2] Escofier, B.; P&agrave;ges, J. (1992) <span style="font-style: italic;">An&aacute;lisis Factoriales Simples y M&uacute;ltiples: Objetivos, M&eacute;todos e Interpretaci&oacute;n.</span> Servicio Editorial Universidad del Pa&iacute;s Vasco, Bilbao.    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[&#160;<a href="javascript:void(0);" onclick="javascript: window.open('/scielo.php?script=sci_nlinks&ref=1943805&pid=S1409-2433201400010000600002&lng=','','width=640,height=500,resizable=yes,scrollbars=1,menubar=yes,');">Links</a>&#160;]<!-- end-ref --></font>    <br> <font style="font-family: Verdana;" size="2"></font>    <!-- ref --><br> <font style="font-family: Verdana;" size="2">[3] Lebart, L.; Morineau, A.; Piron, M. (2006) <span style="font-style: italic;">Statistique Exploratoire Multidimensionnelle</span>. Dunod, Paris.    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[&#160;<a href="javascript:void(0);" onclick="javascript: window.open('/scielo.php?script=sci_nlinks&ref=1943808&pid=S1409-2433201400010000600003&lng=','','width=640,height=500,resizable=yes,scrollbars=1,menubar=yes,');">Links</a>&#160;]<!-- end-ref --></font>    <br> <font style="font-family: Verdana;" size="2"></font>    <!-- ref --><br> <font style="font-family: Verdana;" size="2">[4] Russolillo, G. (2012) &#8220;Non-metric partial least squares&#8221;, <span style="font-style: italic;">Electronic Journal of Statistics</span><span  style="font-weight: bold;"> </span>6: 1648&#8211;1655.    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[&#160;<a href="javascript:void(0);" onclick="javascript: window.open('/scielo.php?script=sci_nlinks&ref=1943811&pid=S1409-2433201400010000600004&lng=','','width=640,height=500,resizable=yes,scrollbars=1,menubar=yes,');">Links</a>&#160;]<!-- end-ref --></font>    <br> <font style="font-family: Verdana;" size="2"></font>    ]]></body>
<body><![CDATA[<!-- ref --><br> <font style="font-family: Verdana;" size="2">[5] Saporta, G. (2011) <span  style="font-style: italic;">Probabilit&eacute;s, Analyse des Donn&eacute;es et Statistique.</span> Editions Technip, Paris.    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[&#160;<a href="javascript:void(0);" onclick="javascript: window.open('/scielo.php?script=sci_nlinks&ref=1943814&pid=S1409-2433201400010000600005&lng=','','width=640,height=500,resizable=yes,scrollbars=1,menubar=yes,');">Links</a>&#160;]<!-- end-ref --></font>    <br> <font style="font-family: Verdana;" size="2"></font>    <!-- ref --><br> <font style="font-family: Verdana;" size="2">[6] Tenenhaus, M. (1998) <span  style="font-style: italic;">La R&eacute;gression PLS. Th&eacute;orie et Pratique.</span> Editions Technip, Paris.    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[&#160;<a href="javascript:void(0);" onclick="javascript: window.open('/scielo.php?script=sci_nlinks&ref=1943817&pid=S1409-2433201400010000600006&lng=','','width=640,height=500,resizable=yes,scrollbars=1,menubar=yes,');">Links</a>&#160;]<!-- end-ref --></font>    <br> <font style="font-family: Verdana;" size="2"></font>    <!-- ref --><br> <font style="font-family: Verdana;" size="2">[7] Wold, H. (1975) &#8220;Path models with latent variables: The non-linear iterative partial least squares (NIPALS) approach&#8221;, Academic Press, New York: 307&#8211;357.    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[&#160;<a href="javascript:void(0);" onclick="javascript: window.open('/scielo.php?script=sci_nlinks&ref=1943820&pid=S1409-2433201400010000600007&lng=','','width=640,height=500,resizable=yes,scrollbars=1,menubar=yes,');">Links</a>&#160;]<!-- end-ref --></font>    <br> <font style="font-family: Verdana;" size="2"></font>    <!-- ref --><br> <font style="font-family: Verdana;" size="2">[8] Young, F. (1981) &#8220;Quantitative analysis of qualitative data&#8221;, <span style="font-style: italic;">Psychometrika</span> 44(4): 357&#8211;388.    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[&#160;<a href="javascript:void(0);" onclick="javascript: window.open('/scielo.php?script=sci_nlinks&ref=1943823&pid=S1409-2433201400010000600008&lng=','','width=640,height=500,resizable=yes,scrollbars=1,menubar=yes,');">Links</a>&#160;]<!-- end-ref --></font>    <br>     <br> <small><span style="font-family: Verdana;"><a name="Correspondencia1"></a><a  href="#Correspondencia2">*</a>Correspondencia a:</span></small>    <br> <font style="font-family: Verdana;" size="2">Tom&aacute;s Aluja: </font><font  style="font-family: Verdana;" size="2">Laboratori de Modelitzaci&oacute; i An&agrave;lisi de la Informaci&oacute; (LIAM), Universitat Polit&egrave;cnica de Catalunya, Barcelona, Espa&ntilde;a. E-Mail: tomas.aluja@upc.edu</font>    <br> <font style="font-family: Verdana;" size="2">V&iacute;ctor Manuel Gonz&aacute;lez: </font><font style="font-family: Verdana;" size="2">Misma direcci&oacute;n que/same address as: Barcelona, Espa&ntilde;a. E-Mail: victor.manuel.gonzalez.rojas@upc.edu</font>    <br> <font style="font-family: Verdana;" size="2"><a name="1"></a><a  href="#3">*</a>Laboratori de Modelitzaci&oacute; i An&agrave;lisi de la Informaci&oacute; (LIAM), Universitat Polit&egrave;cnica de Catalunya, Barcelona, Espa&ntilde;a. E-Mail: tomas.aluja@upc.edu</font>    <br> <font style="font-family: Verdana;" size="2"><a name="2"></a><a  href="#4">&#8224;</a>Misma direcci&oacute;n que/same address as: Barcelona, Espa&ntilde;a. E-Mail: victor.manuel.gonzalez.rojas@upc.edu</font>    <br> <hr style="width: 100%; height: 2px;">     <div style="text-align: center;"><font style="font-family: Verdana;"  size="2"></font><font style="font-family: Verdana; font-weight: bold;"  size="2">Received: 7/May/2013; Revised: 14/Nov/2013; Accepted: 15/Nov/2013 </font></div> </div>      ]]></body><back>
<ref-list>
<ref id="B1">
<label>1</label><nlm-citation citation-type="book">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Aluja]]></surname>
<given-names><![CDATA[T]]></given-names>
</name>
<name>
<surname><![CDATA[Morineau]]></surname>
<given-names><![CDATA[A.]]></given-names>
</name>
</person-group>
<source><![CDATA[Aprender de los Datos: El Análisis de Componentes Principales]]></source>
<year>1999</year>
<publisher-loc><![CDATA[^eBarcelona Barcelona]]></publisher-loc>
<publisher-name><![CDATA[EUB S.L]]></publisher-name>
</nlm-citation>
</ref>
<ref id="B2">
<label>2</label><nlm-citation citation-type="book">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Escofier]]></surname>
<given-names><![CDATA[B.]]></given-names>
</name>
<name>
<surname><![CDATA[Pàges]]></surname>
<given-names><![CDATA[J.]]></given-names>
</name>
</person-group>
<source><![CDATA[Análisis Factoriales Simples y Múltiples: Objetivos, Métodos e Interpretación]]></source>
<year>1992</year>
<publisher-loc><![CDATA[^eBilbao Bilbao]]></publisher-loc>
<publisher-name><![CDATA[Servicio Editorial Universidad del País Vasco]]></publisher-name>
</nlm-citation>
</ref>
<ref id="B3">
<label>3</label><nlm-citation citation-type="book">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Lebart]]></surname>
<given-names><![CDATA[L.]]></given-names>
</name>
<name>
<surname><![CDATA[Morineau]]></surname>
<given-names><![CDATA[A.]]></given-names>
</name>
<name>
<surname><![CDATA[Piron]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
</person-group>
<source><![CDATA[Statistique Exploratoire Multidimensionnelle]]></source>
<year>2006</year>
<publisher-loc><![CDATA[^eParis Paris]]></publisher-loc>
<publisher-name><![CDATA[Dunod]]></publisher-name>
</nlm-citation>
</ref>
<ref id="B4">
<label>4</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Russolillo]]></surname>
<given-names><![CDATA[G.]]></given-names>
</name>
</person-group>
<article-title xml:lang="en"><![CDATA[Non-metric partial least squares]]></article-title>
<source><![CDATA[Electronic Journal of Statistics]]></source>
<year>2012</year>
<volume>6</volume>
<page-range>1648-1655</page-range></nlm-citation>
</ref>
<ref id="B5">
<label>5</label><nlm-citation citation-type="book">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Saporta]]></surname>
<given-names><![CDATA[G.]]></given-names>
</name>
</person-group>
<source><![CDATA[Probabilités, Analyse des Données et Statistique]]></source>
<year>2011</year>
<publisher-loc><![CDATA[^eParis Paris]]></publisher-loc>
<publisher-name><![CDATA[Editions Technip]]></publisher-name>
</nlm-citation>
</ref>
<ref id="B6">
<label>6</label><nlm-citation citation-type="book">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Tenenhaus]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
</person-group>
<source><![CDATA[La Régression PLS.: Théorie et Pratique]]></source>
<year>1998</year>
<publisher-loc><![CDATA[^eParis Paris]]></publisher-loc>
<publisher-name><![CDATA[Editions Technip]]></publisher-name>
</nlm-citation>
</ref>
<ref id="B7">
<label>7</label><nlm-citation citation-type="book">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Wold]]></surname>
<given-names><![CDATA[H.]]></given-names>
</name>
</person-group>
<source><![CDATA[Path models with latent variables: The non-linear iterative partial least squares (NIPALS) approach]]></source>
<year>1975</year>
<page-range>307-357</page-range><publisher-loc><![CDATA[^eNew York New York]]></publisher-loc>
<publisher-name><![CDATA[Academic Press]]></publisher-name>
</nlm-citation>
</ref>
<ref id="B8">
<label>8</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Young]]></surname>
<given-names><![CDATA[F.]]></given-names>
</name>
</person-group>
<article-title xml:lang="en"><![CDATA[Quantitative analysis of qualitative data]]></article-title>
<source><![CDATA[Psychometrika]]></source>
<year>1981</year>
<volume>44</volume>
<numero>4</numero>
<issue>4</issue>
<page-range>357-388</page-range></nlm-citation>
</ref>
</ref-list>
</back>
</article>
