[R-sig-ME] Repeated measures in lmer().

Rolf Turner r.turner at auckland.ac.nz
Thu Apr 2 04:26:50 CEST 2009


On 2/04/2009, at 3:20 PM, Steven McKinney wrote:

> Hi Rolf
>
> E  <- rnorm(4*NCHILD,0,0)
>
> Is this correct?  Did you mean to have variance = 0?

Yes.  The E is just window dressing; it doesn't do anything.
I just put it there to emphasize that measurement error *could*
be there, but I really want to assume it away/absorb it into
the ``child effect'' (since it can't be distinguished from being
part of the child effect without replication).

I *thought* I had put a comment to this effect into my code; guess
it got deleted in the editing process.

Sorry for the confusion.

	cheers,

		Rolf

######################################################################
Attention:\ This e-mail message is privileged and confid...{{dropped:9}}




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