Hi.
I think augPred objects are basically dataframes with a column labeled
".groups". Try:
mypreddat<-augPred(lme.object)
names(mypreddat) ##Just to see what those names are
xyplot(y~x, ##probably "y" and "x" will be different names here
groups=.groups,
subset=.type=="predicted", ##incase you just want predicted values,
panel=function(x,y,groups,...){
panel.superpose(x,y,groups,...)})
This might be more complicated with more than 2 levels of groups. But I
hope it will get you started.
Best,
Jake
------
Jake Bowers
Dept of Political Science
University of Michigan
On Thu, 26 Jun 2003, Kyle Roberts wrote:
> Dear All,
>
> I recently posted a question to the group but got no response, so I will try
> again in another way. If I use the command:
>
> plot(augPred(lme.object))
>
> I get a graph with the fitted regression lines for each level-2 structure
> within a HLM/multilevel design (see Pinheiro and Bates, p. 40). Although
> this is useful, I want to also see a graph where each of these fitted
> regression lines are all in the same box/area, not a different box for each
> group.
>
> Any ideas?
>
> Kyle
>
> *****************************************************
> J. Kyle Roberts, Ph.D.
> Assistant Professor of Educational Research
> University of North Texas
> P.O.Box 311337
> Denton, TX 76203-1337
> voice - 940-369-7168
> fax - 940-565-2185
> kroberts@unt.edu
> *****************************************************
> --------------------------------------------------------------------
> 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
>
>
|