next up previous
Next: About this document ... Up: Dimension Reduction Regression in Previous: References

Function documentation

drDimension reduction regression dimension reduction, inverse regression, regressiondr
\begin{Description}\relax
The function dr implements dimension reduction methods, including SIR, SAVE and pHd.\end{Description}

\begin{Usage}
\begin{verbatim}dr(formula, data=list(), subset, weights, na.act...
...a.omit, method=''sir'',
contrasts=NULL,numdir=4, ...)\end{verbatim}\end{Usage}

\begin{Arguments}
\begin{ldescription}
\item[\code{formula}] a symbolic descript...
...object}]
\end{ldescription} A dr object created by a call to dr.\end{Arguments}

\begin{Details}\relax
The general regression problem studies \eqn{F(y\vert x)}{}...
... default plot
method for dr objects, based on a scatterplot matrix.\end{Details}

\begin{Value}
Unless \code{estimate.weights=T}, returns a list of items, includi...
...ldescription}\par Other returned values repeat quantities from input.\end{Value}

\begin{Author}\relax
Sanford Weisberg,
sandy@stat.umn.edu\end{Author}

\begin{References}\relax
The details of these methods are given by R. D. Cook
(...
... Computing and Graphics, New
York: Wiley, www.stat.umn.edu/arc.\end{References}

\begin{SeeAlso}\relax
dr.permutation.test,dr.x,dr.y,dr.direction,dr.coplot,plot.dr\end{SeeAlso}

\begin{Examples}
\begin{ExampleCode}
library(dr)
data(ais)
attach(ais)  ...
dr.permutation.testInverse Regression Permutation Tests inverse regression, regressiondr.permutation.test
\begin{Description}\relax
This function computes a permutation test for dimension for any inverse
regression fitting method.\end{Description}

\begin{Usage}
\begin{verbatim}dr.permutation.test(object, npermute=50, numdir=object$numdir)\end{verbatim}\end{Usage}

\begin{Arguments}
\begin{ldescription}
\item[\code{object}] an inverse regressio...
...he dimension, with the default from
the object
\end{ldescription}\end{Arguments}

\begin{Value}
Returns an object of type 'dr.permutation.test' that can be printed or
summarized to give the summary of the test.\end{Value}

\begin{Author}\relax
Sanford Weisberg,
sandy@stat.umn.edu\end{Author}

\begin{References}\relax
See www.stat.umn.edu/arc/addons.html, and then select the article
on inverse regression.\end{References}

\begin{SeeAlso}\relax
\code{\Link{dr}}\end{SeeAlso}

\begin{Examples}
\begin{ExampleCode}
data(ais)
attach(ais)  ...
plot.drPlotting methods for dimension reduction regression coplot.drplot.dr dr, dimension reduction, inverse regression, graphicsplot.dr
\begin{Description}\relax
These routines provide default plotting methods for dimension reduction regression.\end{Description}

\begin{Usage}
\begin{verbatim}plot.dr(object, which=1:object$numdir, mark.by.y...
...coplot(object, which=1:object$numdir, mark.by.y=F, ...)\end{verbatim}\end{Usage}

\begin{Arguments}
\begin{ldescription}
\item[\code{object}] Any dimension reduct...
...anel=panel.car will add smoothers
to a coplot.
\end{ldescription}\end{Arguments}

\begin{Value}
Produces a scatterplot matrix (plot) or coplot (dr.coplot) of the specified
directions in an dimension reduction regression\end{Value}

\begin{Author}\relax
Sanford Weisberg,
sandy@stat.umn.edu\end{Author}

\begin{References}\relax
Cook, R. D. and Weisberg, S. (1999). Applied Regression
Including Computing and Graphics. New York: Wiley.\end{References}

\begin{Examples}
\begin{ExampleCode}
...
rotplotdraw many 2D projections of a 3D plot rotplot graphicsrotplot
\begin{Description}\relax
This function draws several 2D views of a 3D plot, sort of like a spinning
plot.\end{Description}

\begin{Usage}
\begin{verbatim}rotplot(x, y, theta=seq(0, pi/2, length = 9), ...)\end{verbatim}\end{Usage}

\begin{Arguments}
\begin{ldescription}
\item[\code{x}] a matrix with 2 columns g...
...de{...}] additional arguments passed to coplot
\end{ldescription}\end{Arguments}

\begin{Details}\relax
For each value of theta, draw the plot of cos(theta)*x[,1]+sin(theta)*x[,2]
versus y.\end{Details}

\begin{Value}
returns a graph object.\end{Value}

\begin{Author}\relax
Sanford Weisberg,
sandy@stat.umn.edu\end{Author}

\begin{Examples}
\begin{ExampleCode}
data(ais)
attach(ais)
m1 <- dr(LBM ~ Ht ...
...rection(m1,which=1:2),dr.y(m1),col=markby(Sex))
\end{ExampleCode}\end{Examples}
dr.xInverse regression term matrix dr.ydr.x dr.zdr.x Dimension reduction regression, regressiondr.x
\begin{Description}\relax
dr.x returns the matrix of data constructed from the f...
...or a
dimension reduction regression. dr.y returns the response.\end{Description}

\begin{Usage}
\begin{verbatim}dr.x(object)
dr.y(object)
dr.z(x,weights=NULL,center=TRUE,rotate=TRUE,decomp=''svd'')\end{verbatim}\end{Usage}

\begin{Arguments}
\begin{ldescription}
\item[\code{object}] An dimension reducti...
...ion to be used in computing the rotation; the
default is ''svd''.\end{Arguments}

\begin{Value}
dr.x returns an \eqn{n \times p}{n by p} matrix of terms, excludin...
...e
response. dr.z returns a possibly centered and scaled version of x.\end{Value}

\begin{Author}\relax
Sanford Weisberg, <sandy@stat.umn.edu>\end{Author}

\begin{SeeAlso}\relax
dr\end{SeeAlso}

\begin{Examples}
\begin{ExampleCode}
data(ais)
attach(ais)
m1 <- dr(LBM~Ht+Wt+RCC+WCC)
dr.x(m1)
dr.y(m1)
\end{ExampleCode}\end{Examples}
dr.estimate.weightsCompute estimated weighting toward normality robust.center.scaledr.estimate.weights Dimension reduction regression, robustnessdr.estimate.weights
\begin{Description}\relax
These functions estimate weights to apply to the rows ...
...weighted matrix as close to multivariate normality as
possible.\end{Description}

\begin{Usage}
\begin{verbatim}dr.weights(formula,...)
dr.estimate.weights(obje...
...mve'', nsamples=NULL, ...)
robust.center.scale(x, ... )\end{verbatim}\end{Usage}

\begin{Arguments}
\begin{ldescription}
\item[\code{object}] a dimension reductio...
...] Additional args are passed to \code{cov.rob}
\end{ldescription}\end{Arguments}

\begin{Details}\relax
The basic outline is: (1) Estimate a mean m and covariance...
... call to \code{dr}, then only the
list of weights will be returned.\end{Details}

\begin{Value}
Returns a list of n weights, some of which may be zero.\end{Value}

\begin{Author}\relax
Sanford Weisberg,
sandy@stat.umn.edu\end{Author}

\begin{References}\relax
R. D. Cook and C. Nachtsheim (1994), Reweighting to ach...
.... Journal of the American
Statistical Association, 89, 592--599.\end{References}

\begin{SeeAlso}\relax
SEE ALSO \code{\Link{lqs}},\code{\Link{rob.cov}}\end{SeeAlso}

\begin{Examples}
\begin{ExampleCode}
\end{ExampleCode}\end{Examples}
The following documentation is from the package lqs by Brian Ripley, and is included here for convenience: cov.robResistant Estimation of Multivariate Location and Scatter cov.mcdcov.rob cov.mvecov.rob robustcov.rob multivariatecov.rob
\begin{Description}\relax
Compute a multivariate location and scale estimate wit...
.... \code{cov.mve} and
\code{cov.mcd} are compatibility wrappers.\end{Description}

\begin{Usage}
\begin{verbatim}cov.rob(x, cor = FALSE, quantile.used = floor((n...
...e.used = floor((n + p + 1)/2),
nsamp = ''best'', seed)\end{verbatim}\end{Usage}

\begin{Arguments}
\begin{ldescription}
\item[\code{x}] a matrix or data frame.
\...
...ndom.seed} will be preserved if it is set.
\par\end{ldescription}\end{Arguments}

\begin{Details}\relax
For method \code{''mve''}, an approximate search is made o...
...ed (as this will have no larger a determinant, and may be smaller).\end{Details}

\begin{Value}
A list with components
\par\begin{ldescription}
\item[\code{center...
...em[\code{n.obs}] total number of observations.
\par\end{ldescription}\end{Value}

\begin{Author}\relax
B.D. Ripley\end{Author}

\begin{References}\relax
P. J. Rousseeuw and A. M. Leroy (1987)
\emph{Robust Re...
...}
ed Y. Dodge, IMS Lecture Notes volume \bold{31}, pp. 201--214.\end{References}

\begin{SeeAlso}\relax
\code{\Link{lqs}}\end{SeeAlso}

\begin{Examples}
\begin{ExampleCode}
data(stackloss)
set.seed(123)
cov.rob(stack...
...ob(stack.x, method = ''mcd'', nsamp = ''exact'')
\end{ExampleCode}\end{Examples}

next up previous
Next: About this document ... Up: Dimension Reduction Regression in Previous: References
Sandy Weisberg 2002-01-10