The following may work:
x <- rnorm(100)
y <- rnorm(100)
z <- runif(100, 0, 5)
plot(x, y, type = 'n')
apply(cbind(x,y,z), 1, function(k) points(k['x'], k['y'], cex=k['z']))
key( x = par('usr')[1], y = par('usr')[3] - par('cxy')[2],
points = list( pch = c(1, 1, 1), cex=c(1, 3, 5)),
text = list( as.character(c(1,3,5) )),
columns = 3,
transparent = TRUE)
--Matt
Matt Austin
Statistician
Amgen
One Amgen Center Drive
M/S 24-2-C
Thousand Oaks CA 93021
(805) 447 - 7431
> -----Original Message-----
> From: s-news-owner@lists.biostat.wustl.edu
> [mailto:s-news-owner@lists.biostat.wustl.edu]On Behalf Of Golubiewski
> Nancy
> Sent: Tuesday, January 25, 2005 19:49 PM
> To: s-news@wubios.wustl.edu
> Subject: [S] legend for graph with "vary symbol size" by column z
> (continuous variable)
>
>
>
>
> Ive plotted a scatter plot with x, y variables, and added information
> about a 3rd variable, z, by varying the symbol size. Its a continuous
> variable, so the typical method for inserting a legend to detail the
> symbol size doesnt work (all examples Ive found in the Online
> Programmers
> manual, Users manual, and menu interface work only when this
> 3rd variable
> is categorical). I could label a few of the data points, but
> Id like to
> indicate how the symbol size varies through a legend. Is it
> possible to
> force the legend to plot a representative symbol for 3 values
> (e.g. 10,
> 100, 1000) in order to provide a scale? Is it possible to
> create a legend
> for this scenario at all?
>
> (Im using Splus 6.1).
>
>
>
> Thanks in advance for your help.
> ~
> (Please reply to N.Golubiewski@massey.ac.nz)
>
> --------------------------------------------------------------------
> This message was distributed by s-news@lists.biostat.wustl.edu. To
> unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
> the BODY of the message: unsubscribe s-news
>
|