[R-sig-ME] nlme

[Catherine Bois]

Thanks guys. I am using lsmeans, however the problem is that I am  
interested in a categorical:continuous predictor interaction so I have  
two problems

1. lsmeans seems to take one value of the continuous predictor and  
compare the categorical predictors in relation to the outcome variable  
on this one value, as opposed to comparing slopes of the variable; I  
therefore wanted to graph this interaction somehow with the adjusted  
means

2. This means that I would need to somehow extract all the model  
adjusted values for these interaction terms and "build my own graph",  
outside of lsmeans capabilities, if that makes sense.

Sorry if this sounds unclear!


Quoting Douglas Bates <bates at stat.wisc.edu> on Wed, 16 Oct 2013  
07:54:16 -0500:

> On Wed, Oct 16, 2013 at 4:13 AM, Catherine Bois <C.Bois at sms.ed.ac.uk> wrote:
>
>> Hi,
>>
>> I am currently using nlme to fit a mixed model. However, I was wondering
>> if there is a function to extract model adjusted means derived from an nlme
>> object, in order to subsequently plot in another programme?
>>
>
> I can't answer your question because I don't know what adjusted means are.
>  I am cc:ing the R-SIG-Mixed-Models at R-project.org mailing list on this
> reply in case someone reading that list can help you.
>



-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.