Re: Using ' fitted' for plots-lme

["F Z" <gerifalte28@hotmail.com>]

You need to tell us better what is your model structure (i.e. did you use a 
Population/individual random effect?).
Anyway, If I understand your question you can use predict.lme() specifying 
you population using the option "Level".  The you can plot the predictions 
vs. observed using plot(pred,fit)
Does that help?

Francisco

> From: "Gupta, Pankaj*" <GuptaP@cder.fda.gov>
> To: "'s-news@lists.biostat.wustl.edu'" <s-news@lists.biostat.wustl.edu>
> Subject: [S] Using ' fitted' for plots-lme
> Date: Thu, 30 Sep 2004 15:47:28 -0400
>
> Dear users,
>
> I wish to generate a plot of the population fitted values vs.observed 
> values
> for a linear mixed effects model. The following gives the individual fitted
> values vs. observed values:
>
> plot(lme.fitfile,response variable~fitted(.))
>
>  Is there a similar way to get the population fitted vs .observed plot?
>
> I'd appreciate any help in this regard.
>
> Thanks
>
> Pankaj
_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today - it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/