[R] Help on eqscplot scatterplot

[Thomas W Blackwell]

H  -

To make a slightly sophisticated plot such as you describe,
most R users would build up the plot from successive calls to
the plotting primitive functions  plot(), points(), lines(), etc.
For a single plot, where you know what the data is, it's routine
to guarantee equal scales by explicitly setting the range covered
by each axis.  Use the arguments xlim= and ylim= in the call to
plot().  You will have to experiment a bit with a ruler to allow
for changes in the aspect ratio by hardcopy printing devices.

Yes, if you want to do something complicated, you do it by trial
and error, adjusting a little bit each time.

-  tom blackwell  -  u michigan medical school  -  ann arbor  -



On Tue, 18 Feb 2003 hgoehlmann at gmx.de wrote:

> Hi,
>
> I am using the equal scales scatterplot plot (eqscplot {MASS}), into which I
> plot a few thousand data points. The data points which accumulate towards
> the centrum of the plot are not informative while the data points furthest away
> from the centrum are most informative. I therefore would like to create a
> combined plot which instead of drawing all data points, would only draw the
> less informative data points via a density surface and starting from a defined
> distance from the centrum the actual data points would be plotted.
>
> Has anyone worked with this kind of problem? I have tried combinations of
> filled.contour with plotting individual points, but I really would need to have
> an geometrically equal scales plot. Should I study the lattice graphs?
>
> I would appreciate any comment you can give.
>
> Cheers,
> hinrich   d8-)
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> http://www.stat.math.ethz.ch/mailman/listinfo/r-help
>