From: Paul Murrell <p.murrell_at_auckland.ac.nz>

Date: Fri, 13 Jun 2008 11:20:00 +1200

*>
*

> ______________________________________________

*> R-help_at_r-project.org mailing list
*

*> https://stat.ethz.ch/mailman/listinfo/r-help
*

*> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
*

*> and provide commented, minimal, self-contained, reproducible code.
*

Date: Fri, 13 Jun 2008 11:20:00 +1200

Hi

Sébastien wrote:

*> Dear R-users,
**>
*

> This is a follow-up on a quite old post of mine which dealt with margins

*> in the pairs function. I thought my problem was solved, but it doesn't
**> seem so (see the code below).
**>
**> I use the pairs function to produce matrix plots, where distinct groups
**> are represented by dots of different colors within each panel. My
**> troubles start when I add a legend at the bottom of the plot. The steps are:
**> 1- I build my dataframe
**> 2- I plot the graph and define the margins within pairs with oma (I want
**> to add 5 + my number of group lines at the bottom)
**> 3- I use gridBase functions to add the legend (symbol + text)
**>
**> The result is an overlay of the legend on the plot. It looks like the
**> lines do not have the same size when using pairs or the gridBase
**> functions. Could anyone explain this discrepancy to me?
*

The problem is that when R (traditional graphics) draws multiple plots on a page it automatically scales down the cex (notice that the axis labels are smaller than normal in the pairs plots). This affects the sizes of margins too. You can see the effect by replacing the pairs() call with just ...

par(mar=c(5 + my.Group.number, 3, 5, 3)) plot(1:10)

... which does leave enough space for your legend.

The downscaling is a factor of .83 if you have 2x2 plots and .66 if you have more than 2 rows or cols of plots, so one solution is to rescale your margins by the corresponding amounts, e.g., ...

oma = 1/.66 * c(5 + my.Group.number, 3, 5, 3)

... which should at least work for your example.

Paul

> Thank you in advance for your help.

*>
**> Sebastien
**>
**>
**> ####### START########
**> library(gridBase)
**>
**> my.Group.number=12
**>
**> my.df<-data.frame(rnorm(my.Group.number*20, mean=5,
**> sd=2),rnorm(my.Group.number*20, mean=4,
**> sd=1.6),rnorm(my.Group.number*20, mean=8,
**> sd=3),rnorm(my.Group.number*20, mean=6,
**> sd=1),rep(letters[1:my.Group.number],each=20))
**> names(my.df)<-c("Var1","Var2","Var2","Var2","Group")
**>
**> pairs(my.df[1:4],
**> main = "My Title",
**> pch = 21,
**> col =rainbow(n=my.Group.number)[unclass(my.df$Group)],
**> bg = rainbow(n=my.Group.number)[unclass(my.df$Group)],
**> oma = c(5 + my.Group.number, 3, 5, 3)
**> )
**>
**> mylegend = paste("Group ",letters[1:my.Group.number],sep="")
**> mylegend.width = strwidth(mylegend[which.max(nchar(mylegend))],
**> "figure", mylegend.Cex(mylegend,myTarget = 0.90))
**>
**> vps <- baseViewports()
**> pushViewport(vps$inner)
**>
**> for (j in 1:my.Group.number) {
**> grid.text(mylegend[j],
**> x = unit((1-mylegend.width)/2,"npc"),
**> y = unit(1 + my.Group.number - j,"lines"),
**> gp = gpar(cex = 1),
**> just = "left",
**> draw=TRUE)
**>
**> grid.points(x =
**> unit((1-mylegend.width)/2,"npc")-convertUnit(unit(0.5,"lines"),"npc"),
**> y = unit(1 + my.Group.number - j,"lines"),
**> size = unit(0.5, "lines"),
**> default.units = "lines",
**> pch = 21,
**> gp = gpar(col = rainbow(n=my.Group.number)[j],fill =
**> rainbow(n=my.Group.number)[j]),
**> draw=TRUE)
**> }
**>
**> ######## END ########
**>
**> Prof Brian Ripley a écrit :
*

>> On Tue, 28 Aug 2007, Sébastien wrote: >> >>> Thanks you very much. >>> I did not thought about calling oma inside the pairs function... >>> >>> Do you know by any chance why 'pairs' behaves differently from >>> 'plot', with regards to the plot dimension ? >> Well, it is an array of figures so why should it be the same? See >> chapter 12 of 'An Introduction to R'. >> >>> Prof Brian Ripley a écrit : >>>>> From ?pairs >>>> The graphical parameter 'oma' will be set by 'pairs.default' >>>> unless supplied as an argument. >>>> >>>> so try >>>> >>>> pairs(iris[1:4], main = "Anderson's Iris Data -- 3 species", pch = 21, >>>> bg = c("red", "green3", "blue")[unclass(iris$Species)], >>>> oma = c(8,3,5,3)) >>>> >>>> >>>> >>>> On Tue, 28 Aug 2007, Sébastien wrote: >>>> >>>>> Dear R-users, >>>>> >>>>> I would like to add a legend at the bottom of pairs plots (it's my >>>>> first >>>>> use of this function). With the plot function, I usually add some >>>>> additional space at the bottom when I define the size of the graphical >>>>> device (using mar); grid functions then allows me to draw my legend >>>>> as I >>>>> want. >>>>> Unfortunatley, this technique does not seem to work with the pairs >>>>> function as the generated plots use all the available space on the >>>>> device (see below). I guess I am missing a key argument... my attempts >>>>> to modify the oma, mar, usr arguments were unsuccesfull, and I >>>>> could not >>>>> find any helpful threads on the archives. >>>>> >>>>> As usual, any advice would be greatly appreciated >>>>> >>>>> Sebastien >>>>> >>>>> >>>>> pdf(file="C:/test.pdf", width=6, height= 6 + 0.2*6) >>>>> >>>>> par(mar=c(5 + 6,4,4,2)+0.1) >>>>> >>>>> pairs(iris[1:4], main = "Anderson's Iris Data -- 3 species", pch = 21, >>>>> bg = c("red", "green3", "blue")[unclass(iris$Species)]) >>>>> >>>>> dev.off() >>>>> >>>>> ______________________________________________ >>>>> R-help_at_stat.math.ethz.ch mailing list >>>>> https://stat.ethz.ch/mailman/listinfo/r-help >>>>> PLEASE do read the posting guide >>>>> http://www.R-project.org/posting-guide.html >>>>> and provide commented, minimal, self-contained, reproducible code. >>>>>

> ______________________________________________

-- Dr Paul Murrell Department of Statistics The University of Auckland Private Bag 92019 Auckland New Zealand 64 9 3737599 x85392 paul_at_stat.auckland.ac.nz http://www.stat.auckland.ac.nz/~paul/ ______________________________________________ R-help_at_r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.Received on Fri 13 Jun 2008 - 00:58:46 GMT

Archive maintained by Robert King, hosted by
the discipline of
statistics at the
University of Newcastle,
Australia.

Archive generated by hypermail 2.2.0, at Fri 13 Jun 2008 - 01:31:12 GMT.

*
Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help.
Please read the posting
guide before posting to the list.
*