Re: [R] zero random effect sizes with binomial lmer

From: Daniel Ezra Johnson <johnson4_at_babel.ling.upenn.edu>
Date: Tue 02 Jan 2007 - 01:59:55 GMT


> From: Andrew Robinson <A.Robinson_at_ms.unimelb.edu.au>
> Date: Mon 01 Jan 2007 - 19:19:29 GMT
>
> I tried an earlier version of R, on a different platform, and got quite
> different results. Sadly, the *earlier* results are the ones that make
> sense.

Andrew, I tried installing R 2.3.1, it seemed to work as you said, but I realized it was because the default method was PQL instead of Laplace. Switching the method to Laplace, the zero effect came back.

It seems like there's no easy way around this. I would like to compare random effect sizes between different subsets of my data, but this may not be possible using the current functions...

Daniel



R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Tue Jan 02 13:04:43 2007

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Tue 02 Jan 2007 - 02:30:28 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.