📄 kernelpls.fit.html
字号:
<html><head><title>R: Kernel PLS (Dayal and MacGregor)</title>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../../R.css">
</head><body>
<table width="100%" summary="page for kernelpls.fit {pls}"><tr><td>kernelpls.fit {pls}</td><td align="right">R Documentation</td></tr></table>
<h2>Kernel PLS (Dayal and MacGregor)</h2>
<h3>Description</h3>
<p>
Fits a PLSR model with the kernel algorithm.
</p>
<h3>Usage</h3>
<pre>kernelpls.fit(X, Y, ncomp, stripped = FALSE, ...)</pre>
<h3>Arguments</h3>
<table summary="R argblock">
<tr valign="top"><td><code>X</code></td>
<td>
a matrix of observations. <code>NA</code>s and <code>Inf</code>s are not
allowed.</td></tr>
<tr valign="top"><td><code>Y</code></td>
<td>
a vector or matrix of responses. <code>NA</code>s and <code>Inf</code>s
are not allowed.</td></tr>
<tr valign="top"><td><code>ncomp</code></td>
<td>
the number of components to be used in the
modelling.</td></tr>
<tr valign="top"><td><code>stripped</code></td>
<td>
logical. If <code>TRUE</code> the calculations are stripped
as much as possible for speed; this is meant for use with
cross-validation or simulations when only the coefficients are
needed. Defaults to <code>FALSE</code>.</td></tr>
<tr valign="top"><td><code>...</code></td>
<td>
other arguments. Currently ignored.</td></tr>
</table>
<h3>Details</h3>
<p>
This function should not be called directly, but through
the generic functions <code>plsr</code> or <code>mvr</code> with the argument
<code>method="kernelpls"</code> (default). Kernel PLS is particularly efficient
when the number of objects is (much) larger than the number of
variables. The results are equal to the NIPALS algorithm. Several
different forms of kernel PLS have been described in literature, e.g.
by De Jong and Ter Braak, and two algorithms by Dayal and
MacGregor. This function implements the
fastest of the latter, not calculating the crossproduct matrix of
X. In the Dyal & MacGregor paper, this is “algorithm 1”.
</p>
<h3>Value</h3>
<p>
A list containing the following components is returned:
</p>
<table summary="R argblock">
<tr valign="top"><td><code>coefficients</code></td>
<td>
an array of regression coefficients for 1, ...,
<code>ncomp</code> components. The dimensions of <code>coefficients</code> are
<code>c(nvar, npred, ncomp)</code> with <code>nvar</code> the number
of <code>X</code> variables and <code>npred</code> the number of variables to be
predicted in <code>Y</code>.</td></tr>
<tr valign="top"><td><code>scores</code></td>
<td>
a matrix of scores.</td></tr>
<tr valign="top"><td><code>loadings</code></td>
<td>
a matrix of loadings.</td></tr>
<tr valign="top"><td><code>Yscores</code></td>
<td>
a matrix of Y-scores.</td></tr>
<tr valign="top"><td><code>Yloadings</code></td>
<td>
a matrix of Y-loadings.</td></tr>
<tr valign="top"><td><code>projection</code></td>
<td>
the projection matrix used to convert X to scores.</td></tr>
<tr valign="top"><td><code>Xmeans</code></td>
<td>
a vector of means of the X variables.</td></tr>
<tr valign="top"><td><code>Ymeans</code></td>
<td>
a vector of means of the Y variables.</td></tr>
<tr valign="top"><td><code>fitted.values</code></td>
<td>
an array of fitted values. The dimensions of
<code>fitted.values</code> are <code>c(nobj, npred, ncomp)</code> with
<code>nobj</code> the number samples and <code>npred</code> the number of
Y variables.</td></tr>
<tr valign="top"><td><code>residuals</code></td>
<td>
an array of regression residuals. It has the same
dimensions as <code>fitted.values</code>.</td></tr>
<tr valign="top"><td><code>Xvar</code></td>
<td>
a vector with the amount of X-variance explained by each
number of components.</td></tr>
<tr valign="top"><td><code>Xtotvar</code></td>
<td>
Total variance in <code>X</code>.</td></tr>
</table>
<p>
<br>
If <code>stripped</code> is <code>TRUE</code>, only the components
<code>coefficients</code>, <code>Xmeans</code> and <code>Ymeans</code> are returned.</p>
<h3>Author(s)</h3>
<p>
Ron Wehrens and Bj鴕n-Helge Mevik
</p>
<h3>References</h3>
<p>
de Jong, S. and ter Braak, C. J. F. (1994) Comments on the PLS kernel
algorithm. <EM>Journal of Chemometrics</EM>, <B>8</B>, 169–174.
</p>
<p>
Dayal, B. S. and MacGregor, J. F. (1997) Improved PLS algorithms.
<EM>Journal of Chemometrics</EM>, <B>11</B>, 73–85.
</p>
<h3>See Also</h3>
<p>
<code><a href="mvr.html">mvr</a></code>
<code><a href="mvr.html">plsr</a></code>
<code><a href="mvr.html">pcr</a></code>
<code><a href="simpls.fit.html">simpls.fit</a></code>
<code><a href="oscorespls.fit.html">oscorespls.fit</a></code>
</p>
<hr><div align="center">[Package <em>pls</em> version 1.1-0 <a href="00Index.html">Index]</a></div>
</body></html>
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -