R help - Apr 2012 - how to plot separate lm ablines on the same xyplot by group

Hi,
I am trying to use xyplot to plot the relationship between size and day
(y~x) by a food factor that has two levels, low and high. I have 3 reps per
factor/day.  I want the plots from each food treatment on the same axiss,
so I used this code:

xyplot(Size ~ Day, groups = Food, data = louis.data.means,col=1,
pch=c(1,17),
panel=function(x,y,groups,...){
panel.superpose(x,y,groups,...)
tmp.lm<-lm(y~x)
panel.abline(tmp.lm)
panel.text(2, 250, label=format(tmp.lm$coefficients[2], digits=4), pos=4)
}
)
This produces a graph of the two treatments (open circles for the low food
 vs triangles for the high food) on the same plot, but only one regression
line (and slope)  which seems to splits the difference between the two
factors (treats them as the same data set).  I would like to produce a
separate regression line for the data from each of the two factors (high
food vs low food).

Is there a way to subset the lm by the factor "food"?

Louis

	[[alternative HTML version deleted]]

Please read ?panel.superpose  again and pay attention to the
panel.groups argument, where this can be specified.

-- Bert

On Wed, Apr 18, 2012 at 12:34 PM, Louis Plough <lplough at usc.edu>
wrote:
> Hi,
> I am trying to use xyplot to plot the relationship between size and day
> (y~x) by a food factor that has two levels, low and high. I have 3 reps per
> factor/day. ?I want the plots from each food treatment on the same axiss,
> so I used this code:
>
> xyplot(Size ~ Day, groups = Food, data = louis.data.means,col=1,
> pch=c(1,17),
> panel=function(x,y,groups,...){
> panel.superpose(x,y,groups,...)
> tmp.lm<-lm(y~x)
> panel.abline(tmp.lm)
> panel.text(2, 250, label=format(tmp.lm$coefficients[2], digits=4), pos=4)
> }
> )
> This produces a graph of the two treatments (open circles for the low food
> ?vs triangles for the high food) on the same plot, but only one regression
> line (and slope) ?which seems to splits the difference between the two
> factors (treats them as the same data set). ?I would like to produce a
> separate regression line for the data from each of the two factors (high
> food vs low food).
>
> Is there a way to subset the lm by the factor "food"?
>
> Louis
>
> ? ? ? ?[[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.


-- 

Bert Gunter
Genentech Nonclinical Biostatistics

Internal Contact Info:
Phone: 467-7374
Website:
http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm

If you could lead me to an example with code, that would help me figure out
how to do it for my function


On Wed, Apr 18, 2012 at 3:57 PM, Louis Plough <lplough@usc.edu> wrote:
> I read it, but I guess I don't quite understand which arguments to pass
> panel.groups to get different lm objects based on the two groups within
> Food.
>
>
>
> On Wed, Apr 18, 2012 at 3:44 PM, Bert Gunter
<gunter.berton@gene.com>wrote:
>
>> Please read ?panel.superpose  again and pay attention to the
>> panel.groups argument, where this can be specified.
>>
>> -- Bert
>>
>> On Wed, Apr 18, 2012 at 12:34 PM, Louis Plough <lplough@usc.edu>
wrote:
>> > Hi,
>> > I am trying to use xyplot to plot the relationship between size
and day
>> > (y~x) by a food factor that has two levels, low and high. I have 3
reps
>> per
>> > factor/day.  I want the plots from each food treatment on the same
>> axiss,
>> > so I used this code:
>> >
>> > xyplot(Size ~ Day, groups = Food, data = louis.data.means,col=1,
>> > pch=c(1,17),
>> > panel=function(x,y,groups,...){
>> > panel.superpose(x,y,groups,...)
>> > tmp.lm<-lm(y~x)
>> > panel.abline(tmp.lm)
>> > panel.text(2, 250, label=format(tmp.lm$coefficients[2], digits=4),
>> pos=4)
>> > }
>> > )
>> > This produces a graph of the two treatments (open circles for the
low
>> food
>> >  vs triangles for the high food) on the same plot, but only one
>> regression
>> > line (and slope)  which seems to splits the difference between the
two
>> > factors (treats them as the same data set).  I would like to
produce a
>> > separate regression line for the data from each of the two factors
(high
>> > food vs low food).
>> >
>> > Is there a way to subset the lm by the factor "food"?
>> >
>> > Louis
>> >
>> >        [[alternative HTML version deleted]]
>> >
>> > ______________________________________________
>> > R-help@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.
>>
>>
>>
>> --
>>
>> Bert Gunter
>> Genentech Nonclinical Biostatistics
>>
>> Internal Contact Info:
>> Phone: 467-7374
>> Website:
>>
>>
http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm
>>
>
>
	[[alternative HTML version deleted]]

please look at the ancova function in the HH package.


On Wed, Apr 18, 2012 at 3:34 PM, Louis Plough <lplough@usc.edu> wrote:
> Hi,
> I am trying to use xyplot to plot the relationship between size and day
> (y~x) by a food factor that has two levels, low and high. I have 3 reps per
> factor/day.  I want the plots from each food treatment on the same axiss,
> so I used this code:
>
> xyplot(Size ~ Day, groups = Food, data = louis.data.means,col=1,
> pch=c(1,17),
> panel=function(x,y,groups,...){
> panel.superpose(x,y,groups,...)
> tmp.lm<-lm(y~x)
> panel.abline(tmp.lm)
> panel.text(2, 250, label=format(tmp.lm$coefficients[2], digits=4), pos=4)
> }
> )
> This produces a graph of the two treatments (open circles for the low food
>  vs triangles for the high food) on the same plot, but only one regression
> line (and slope)  which seems to splits the difference between the two
> factors (treats them as the same data set).  I would like to produce a
> separate regression line for the data from each of the two factors (high
> food vs low food).
>
> Is there a way to subset the lm by the factor "food"?
>
> Louis
>
>        [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@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.
>
	[[alternative HTML version deleted]]