lme output

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

lme output

John Logsdon-2
I am trying to extract the solution from a simple lme calculation.

For example (the first 4 have a mean 0, sd 1):

> y<-c(-1.118,-.5,.5,1.118,10)
> gp<-factor(c(rep('one',4),'two'))
> res<-lme(y~1,rand=~1|gp)
Linear mixed-effects model fit by REML
  Data: NULL
  Log-restricted-likelihood: -8.67141
  Fixed: y ~ 1
(Intercept)
   4.962502

Random effects:
 Formula: ~1 | gp
        (Intercept)  Residual
StdDev:    7.026737 0.9999747

Number of Observations: 5
Number of Groups: 2

The problem is to get the random effects intercept - essentially the
between groups standard deviation - into a scalar.  I can't see anything
in the values from res.  The other values - fixed intercept and residual -
are clearly available - eg res$coeff$fixed is straight forward.

I would use lmer but this does not appear to return any values.

Can someone advise please?  I do recall having this problem some time ago
but that was with a much older version.  This is with R 2.2.1 and the
latest lme.

TIA


Best wishes

John

John Logsdon                               "Try to make things as simple
Quantex Research Ltd, Manchester UK         as possible but not simpler"
[hidden email]              [hidden email]
+44(0)161 445 4951/G:+44(0)7717758675       www.quantex-research.com

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: lme output

Dieter Menne
John Logsdon <j.logsdon <at> quantex-research.com> writes:

>
> I am trying to extract the solution from a simple lme calculation.
>
> > y<-c(-1.118,-.5,.5,1.118,10)
> > gp<-factor(c(rep('one',4),'two'))
> > res<-lme(y~1,rand=~1|gp)
...
> Random effects:
>  Formula: ~1 | gp
>         (Intercept)  Residual
> StdDev:    7.026737 0.9999747
>
> The problem is to get the random effects intercept - essentially the
> between groups standard deviation - into a scalar.  

VarCorr(res)

will give you the matrix with the relevant numbers

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html