[R-sig-ME] Another case of -1.0 correlation of random effects

Kevin E. Thorpe kevin.thorpe at utoronto.ca
Fri Apr 16 17:31:55 CEST 2010


Andrew Dolman wrote:
> No problem,
> 
> lme1 <- lme(iAUC~Treatment+I(Dose-8), 
> random=~Treatment+I(Dose-8)|Subject,data=gluc)
> summary(lme1)
> 
> works fine for me.
> 
> Your problem might have been that you had lmer and nlme loaded at the 
> same time, or it might just have been the spelling mistake (+Does)
> 
> 
> Andy.

Smacks head!

It was a spelling mistake, just not the one in my email.  I spelled lme 
as lmer.  Clearly, that won't work! :-)

Thanks again.

> 
> 
> andydolman at gmail.com <mailto:andydolman at gmail.com>
> 
> 
>     Maybe, since the data are balanced, I could use lme(), but I have no
>     clue how to specify this random effect structure there.  I tried
>     random=~Treatment+Does|Subject but lme() didn't like that.
> 
>     Thanks for listening.  I appreciate all you've done already.
> 
>     Kevin
> 


-- 
Kevin E. Thorpe
Biostatistician/Trialist, Knowledge Translation Program
Assistant Professor, Dalla Lana School of Public Health
University of Toronto
email: kevin.thorpe at utoronto.ca  Tel: 416.864.5776  Fax: 416.864.3016




More information about the R-sig-mixed-models mailing list