Information Technology Reference
In-Depth Information
The overall quality of the display in terms of
the original space.
CVA.quality.origvar
The axis adequacies.
adequacy
Predictions of the means or of the samples as
specified.
predictions
Matrix containing the canonical means.
Z.means.mat
CVAbipl
is mainly called for its side effect, the plotting of the CVA biplot. The functions
CVA.predictions.mat
and
CVA.predictivities
provide extensive output of the
various measures of fit and fitted values discussed in this chapter.
4.7.2 Function
CVAbipl.zoom
A function for zooming into an interactively selected area of a one- or two-dimensional
CVA biplot. In addition to the arguments of
CVAbipl
, this function has the argument
zoomval
for controlling the amount of zooming. Its usage is similar to that described
for the function
PCAbipl.zoom
in Section 3.7.2.
4.7.3 Function
CVAbipl.density
A function for overlaying two-dimensional CVA biplots on a density plot of a two-
dimensional CVA approximation of the input matrix. Its usage and arguments additional
to those of
CVAbipl
are identical to the description given for
PCAbipl.density
in
Section 3.7.3.
4.7.4 Function
CVAbipl.density.zoom
Analogous to
CVAbipl.zoom
, the function
CVAbipl.density.zoom
provides zooming
functionality for
CVAbipl.density
through its argument
zoomval
.
4.7.5 Function
CVAbipl.pred.regions
This function constructs nearest canonical means regions for a CVA biplot. In addition
to the arguments of
CVAbipl
, it has the arguments described below. These latter
arguments are passed to the function
pred.regions
which in turn calls the function
Colour.Pixel
. The functions
pred.regions
and
Colour.Pixel
are responsible
for calculating the distance from a given grid point in the biplot space to each
canonical mean in the full canonical space. The grid point is then coloured according
to the canonical mean having the shortest distance to it. Additional arguments of
CVAbipl.pred.regions
for passing to functions
pred.regions
and
Colour.Pixel
are as follows: