[R-sig-ME] forcing uncorrelated random effects

Doran, Harold HDoran at air.org
Thu Jun 11 20:19:56 CEST 2009


Here is an example. The second has no correlation.

(fm1 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy))
(fm2 <- lmer(Reaction ~ Days + (Days - 1|Subject), sleepstudy)) 

> -----Original Message-----
> From: r-sig-mixed-models-bounces at r-project.org 
> [mailto:r-sig-mixed-models-bounces at r-project.org] On Behalf 
> Of Claus Wilke
> Sent: Thursday, June 11, 2009 1:58 PM
> To: r-sig-mixed-models at r-project.org
> Subject: [R-sig-ME] forcing uncorrelated random effects
> 
> Dear list,
> 
> by default, lmer assumes that random effects are correlated. 
> Is it possible to force them to be uncorrelated?
> 
> Specifically, assume I'm measuring cell counts in multiple 
> patients over time, and want to fit the following two models:
> > m1 = lmer( count ~ (1|patient)+time )
> > m2 = lmer( count ~ (time|patient)+time )
> Model 2 has two additional parameters over model 1, a 
> variance of random slopes and a covariance of random slopes 
> and random intercepts. How do I specify a model that has 
> random slopes but no covariance between slopes and intercepts?
> 
> Thanks a lot,
>   Claus
> 
> --
> Claus Wilke
> Section of Integrative Biology
>  and Center for Computational Biology and Bioinformatics 
> University of Texas at Austin
> 1 University Station C0930
> Austin, TX 78712
> cwilke at mail.utexas.edu
> 512 471 6028
> 
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list 
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
> 




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