bpCent3d: Three Dimensional Biplot of a Clustering Solution
In BCA: Business and Customer Analytics

Description Usage Arguments Details Side Effects See Also

Plot a biplot of a clustering solution on the current graphics device.

bpCent3d(pc, clsAsgn, data.pts = TRUE, centroids = TRUE, 
  choices = 1:3, scale = 1, pc.biplot = FALSE, var.axes = TRUE, 
  col.score = "black", col.cntrs = "blue", col.load = "red",
  xlabs = NULL, ylabs = NULL, xlim = NULL, ylim = NULL, zlim = NULL,
  xlab = NULL, ylab = NULL, dim.lab = NULL, fov = 60)

`pc`	The `prcomp` object of the data used in clustering.
`clsAsgn`	A vector containing the cluster assignment for each record in the clustering data.
`data.pts`	If `TRUE` the point for each record is plotted.
`centroids`	If `TRUE` the centroid for each cluster is plotted.
`choices`	length 2 vector specifying the components to plot.
`scale`	The variables scaled by `lambda ^ scale` and the observations are scaled by `lambda ^ (1-scale)`, where `lambda` are the eigen values of the principal components solution. `scale` should be between 0 and 1.
`pc.biplot`	If `TRUE`, then `lambda = 1` and the observations are scaled up the sqrt(n) and the variables scaled down by sqrt(n). In this case the inner product between variables approximate covariances, and the distances between observations approximate Mahalanobis distance. Gabriel refers to this as a "principal component biplot".
`var.axes`	If `TRUE` the second set of points have arrows representing them as (unscaled) axes.
`col.score, col.cntrs, col.load`	A vector of length 2 giving the colours for the first and second set of points respectively (and the corresponding axes). If a single colour is specified it will be used for both sets. If missing the default colour is looked for in the `palette`: if there it and the next colour as used, otherwise the first two colours of the paletter are used.
`xlabs`	A vector of character strings to label the first set of points: the default is to use the row dimname of `x`, or `1:n` is the dimname is `NULL`.
`ylabs`	A vector of character strings to label the second set of points: the default is to use the row dimname of `y`, or `1:n` is the dimname is `NULL`.
`xlim, ylim, zlim`	Limits for the x, y, z axes in the units of the first set of variables.
`xlab, ylab, dim.lab, fov`	Graphical parameters for rgl.

This function is a reworking of biplot in order to allow the cluster centroids of a clustering solution to be displayed as well as the individual data points, and extend it to three dimensions. The use of arrow heads is discouraged since their creation is very resource intensive. If both data.pts and centroids are set to FALSE then only the variable directional vectors are displayed.

A plot is produced on the current graphics device.

biplot

BCA documentation built on May 2, 2019, 1:26 p.m.