R help - Jul 2011 - How to adjust y-axis when using panel.densityplot within histogram function

Hi

I would like to superimpose group-specific densityplots on top of an overall
histogram using panel.histogram and panel.densityplot. Furthermore, I would
like to automatically adjust the range of the y-axis to take into account
the ranges of both histogram and densityplot. This last part is where I have
a problem. I believe using the prepanel argument of histogram is typically
the way to go, but I did not have very much success with my attempts. Is
this the way to go, or should I calculate the range outside the histogram
call using some stat function (density?) on each group and use it as a ylim
argument?

Any advise would be greatly appreciated.

Thank you

Sebastien

require(lattice)
set.seed(12345)

foo1 <- data.frame(x=rnorm(100,0,0.1),grp=1)
foo2 <- data.frame(x=rnorm(100,2,1),grp=2)

foo <- rbind(foo1,foo2)

#
# Some code to calculate ylim...
#

histogram(~x,data=foo,groups=grp,
  #prepanel=function(x,type,groups,...){???},
  panel=function(x,type,groups,...){
    panel.histogram(x,col='transparent',...)
    panel.densityplot(x,groups=groups,...)
  },
  #ylim=ylim,
  type='density')

	[[alternative HTML version deleted]]

On Jul 27, 2011, at 9:15 AM, S?bastien Bihorel wrote:
> Hi
>
> I would like to superimpose group-specific densityplots on top of an  
> overall
> histogram using panel.histogram and panel.densityplot. Furthermore,  
> I would
> like to automatically adjust the range of the y-axis to take into  
> account
> the ranges of both histogram and densityplot. This last part is  
> where I have
> a problem. I believe using the prepanel argument of histogram is  
> typically
> the way to go, but I did not have very much success with my  
> attempts. Is
> this the way to go, or should I calculate the range outside the  
> histogram
> call using some stat function (density?) on each group and use it as  
> a ylim
> argument?
>
> Any advise would be greatly appreciated.
>
> Thank you
>
> Sebastien
>
> require(lattice)
> set.seed(12345)
>
> foo1 <- data.frame(x=rnorm(100,0,0.1),grp=1)
> foo2 <- data.frame(x=rnorm(100,2,1),grp=2)
>
> foo <- rbind(foo1,foo2)
>
> #
> # Some code to calculate ylim...
> #
>
> histogram(~x,data=foo,groups=grp,
>  #prepanel=function(x,type,groups,...){???},
>  panel=function(x,type,groups,...){
>    panel.histogram(x,col='transparent',...)
>    panel.densityplot(x,groups=groups,...)
>  },
>  #ylim=ylim,
>  type='density')
Use densityplot instead of histogram as the wrapping function so its  
more extreme ranges are respected. You do get an error when you do  
that saying that 'breaks' is invalid, but if you read the ?histogram  
page, it suggests that setting breaks=NULL might produce acceptable  
default behavior, and that seems to be so in this case:

densityplot(~x,data=foo,groups=grp,
  #prepanel=function(x,type,groups,...){???},
  panel=function(x,type,groups,...){
          panel.densityplot(x,groups=groups,...)
          panel.histogram(x,col='transparent', breaks = NULL, ...)

  } )
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> 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.
David Winsemius, MD
West Hartford, CT