R help - Apr 2009 - Multiple Hexbinplots in 2 columns with a Single Categorical Variable

Dear Ladies and Gentlemen,

 

I have a fairly large database (N=13,000) and a single main categorical
discriminator between the groups.

 

I want to look at the time course of a number of continuous biochemical
variables over chronologic age.

 

Therefore I believe I need to prepare hexbinplots in two columns with simple
regression lines in them (with useOuterStrips (in library(latticeExtra) if
possible) and also single hexbinplots with two regression lines
(panel.lmlines) corresponding to the two groups to facilitate data
comparison.

 

Tick marks and labels should be off in most panels, and the scale in each
row will be different.

 

I have some code written for the NHANES dataset in library(hexbin) which
almost does this job, but leaves blank paper up the middle of the two
columns of panels.

 

I have been exploring lattice for this which almost gets me there but not
quite.

 

The code I have written is as follows.

 

I would be ever so grateful for any advice anyone may be able to offer.

 

With best wishes,

 

Stuart Reece,

Australia.

 

 

useOuterStrips(hexbinplot(Transferin ~ Age | factor(Race) + factor(Sex),
data = NHANES, type = "r",

           aspect = 1, 

           scales = 

           list(x 
                 list(relation = "free", rot = 0,

                 at=list(TRUE, TRUE, NULL, NULL)),

                y = 

                 list(relation = "free", rot = 0,

                 at=list(TRUE, NULL, TRUE, NULL))),

       par.settings        list(layout.heights = list(axis.panel = rep(c(1,
0),c(1, 1)))),
       par.strip.text = list(cex = 1)))          
 

 


	[[alternative HTML version deleted]]

Dear Ladies and Gentlemen,

 I have a fairly large database (N=13,000) and a single main categorical
discriminator between the groups.

 I want to look at the time course of a number of continuous biochemical
variables over chronologic age.

 Therefore I believe I need to prepare hexbinplots in two columns with simple
regression lines in them (with useOuterStrips (in library(latticeExtra) if
possible) and also single hexbinplots with two regression lines (panel.lmlines)
corresponding to the two groups to facilitate data comparison.

 Tick marks and labels should be off in most panels, and the scale in each row
will be different.

 I have some code written for the NHANES dataset in library(hexbin) which almost
does this job, but leaves blank paper up the middle of the two columns of
panels.

 I have been exploring lattice for this which almost gets me there but not
quite.

 The code I have written is as follows.

 I would be ever so grateful for any advice anyone may be able to offer.

 With best wishes,

 Stuart Reece,

Australia.

 

 

useOuterStrips(hexbinplot(Transferin ~ Age | factor(Race) + factor(Sex), data =
NHANES, type = "r",

           aspect = 1, 

           scales = 

           list(x 
                 list(relation = "free", rot = 0,

                 at=list(TRUE, TRUE, NULL, NULL)),

                y = 

                 list(relation = "free", rot = 0,

                 at=list(TRUE, NULL, TRUE, NULL))),

       par.settings =       list(layout.heights = list(axis.panel = rep(c(1,
0),c(1, 1)))),       par.strip.text = list(cex = 1)))

	[[alternative HTML version deleted]]

Hi

Does this do it ... ?

useOuterStrips(hexbinplot(Transferin ~ Age | factor(Race) + factor(Sex),
                           data=NHANES, type="r",
                           aspect=1,
                           scales                          
list(x=list(relation="free", rot=0,
                                  at=list(TRUE, TRUE, NULL, NULL)),
                                y=list(relation="free", rot=0,
                                  at=list(TRUE, NULL, TRUE, NULL))),
                           par.settings                          
list(layout.heights=list(axis.panel=c(1, 0)),
                                layout.widths=list(axis.panel=c(1, 0)),
                                par.strip.text=list(cex=1))))

In this example, the 'relation="free"' is not doing much, so a
simpler
version would be just ...

useOuterStrips(hexbinplot(Transferin ~ Age | factor(Race) + factor(Sex),
                           data = NHANES, type = "r",
                           aspect = 1,
                           scales = list(alternating=1, tck=c(1, 0))))

... but it sounds like that might not be appropriate for your real problem.

Paul


Dr Stuart Reece wrote:
> Dear Ladies and Gentlemen,
> 
>  
> 
> I have a fairly large database (N=13,000) and a single main categorical
> discriminator between the groups.
> 
>  
> 
> I want to look at the time course of a number of continuous biochemical
> variables over chronologic age.
> 
>  
> 
> Therefore I believe I need to prepare hexbinplots in two columns with
simple
> regression lines in them (with useOuterStrips (in library(latticeExtra) if
> possible) and also single hexbinplots with two regression lines
> (panel.lmlines) corresponding to the two groups to facilitate data
> comparison.
> 
>  
> 
> Tick marks and labels should be off in most panels, and the scale in each
> row will be different.
> 
>  
> 
> I have some code written for the NHANES dataset in library(hexbin) which
> almost does this job, but leaves blank paper up the middle of the two
> columns of panels.
> 
>  
> 
> I have been exploring lattice for this which almost gets me there but not
> quite.
> 
>  
> 
> The code I have written is as follows.
> 
>  
> 
> I would be ever so grateful for any advice anyone may be able to offer.
> 
>  
> 
> With best wishes,
> 
>  
> 
> Stuart Reece,
> 
> Australia.
> 
>  
> 
>  
> 
> useOuterStrips(hexbinplot(Transferin ~ Age | factor(Race) + factor(Sex),
> data = NHANES, type = "r",
> 
>            aspect = 1, 
> 
>            scales = 
> 
>            list(x > 
>                  list(relation = "free", rot = 0,
> 
>                  at=list(TRUE, TRUE, NULL, NULL)),
> 
>                 y = 
> 
>                  list(relation = "free", rot = 0,
> 
>                  at=list(TRUE, NULL, TRUE, NULL))),
> 
>        par.settings >        list(layout.heights = list(axis.panel =
rep(c(1, 0),c(1, 1)))),
>        par.strip.text = list(cex = 1)))          
>  
> 
>  
> 
> 
> 	[[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.
-- 
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/

On Thu, Apr 9, 2009 at 4:18 AM, Stuart Reece <sreece at bigpond.net.au>
wrote:
> Dear Ladies and Gentlemen,
>
> ?I have a fairly large database (N=13,000) and a single main categorical
discriminator between the groups.
>
> ?I want to look at the time course of a number of continuous biochemical
variables over chronologic age.
>
> ?Therefore I believe I need to prepare hexbinplots in two columns with
simple regression lines in them (with useOuterStrips (in library(latticeExtra)
if possible) and also single hexbinplots with two regression lines
(panel.lmlines) corresponding to the two groups to facilitate data comparison.
>
> ?Tick marks and labels should be off in most panels, and the scale in each
row will be different.
>
> ?I have some code written for the NHANES dataset in library(hexbin) which
almost does this job, but leaves blank paper up the middle of the two columns of
panels.
>
> ?I have been exploring lattice for this which almost gets me there but not
quite.
>
> ?The code I have written is as follows.
>
> ?I would be ever so grateful for any advice anyone may be able to offer.
>
> ?With best wishes,
>
> ?Stuart Reece,
>
> Australia.
>
>
>
>
>
> useOuterStrips(hexbinplot(Transferin ~ Age | factor(Race) + factor(Sex),
data = NHANES, type = "r",
>
> ? ? ? ? ? aspect = 1,
>
> ? ? ? ? ? scales >
> ? ? ? ? ? list(x >
> ? ? ? ? ? ? ? ? list(relation = "free", rot = 0,
>
> ? ? ? ? ? ? ? ? at=list(TRUE, TRUE, NULL, NULL)),
>
> ? ? ? ? ? ? ? ?y >
> ? ? ? ? ? ? ? ? list(relation = "free", rot = 0,
>
> ? ? ? ? ? ? ? ? at=list(TRUE, NULL, TRUE, NULL))),
>
> ? ? ? par.settings = ? ? ? list(layout.heights = list(axis.panel = rep(c(1,
0),c(1, 1)))),
You need the same trick for the widths:

                          par.settings                          
list(layout.heights = list(axis.panel rep(c(1, 0),c(1, 1))),
                               layout.widths = list(axis.panel rep(c(1, 0),c(1,
1)))),
> par.strip.text = list(cex = 1)))
-Deepayan