Important: The stats package has been deprecated. Use the superseding package Statistics instead.

The function scatterplot with the format parameter format=agglomerated of the subpackage stats[statplots] organizes data clusters into classes.

This type of plot is often used by cartographers, or when there is a large number of data points involved.  The idea is that it is not necessary to have the detail of each individual point in a plot.  Closely grouped points are plotted instead as one box.

When n or more points occur within a cube with side length l those points are replaced by the tightest fitting box possible.  The tightest fitting box in one-dimension will always be a line.  In higher dimensions, rectangles or boxes may be plotted.

When  or more points occur within the same cube described in the last paragraph, the box is emphasized by being plotted with thicker lines.

If n or l are zero, or unspecified, default values will be used.  The default value of n is the cube root of the number of points, divided by the number of dimensions of the data (that is, number of statistical lists passed as parameters).  The default value for l is one-tenth the range of the x-coordinate data.

Class data is converted to classmarks before generating the plot.  Weighted data is accounted for.  Missing data is ignored.

The command with(stats[statplots]) allows the use of the abbreviated form of this command.