[R-sig-ME] No estimated scale value given

[Renwick, A. R.]

This has just recently been discussed:
https://stat.ethz.ch/pipermail/r-sig-mixed-models/2009q1/001922.html
Hope that help,
Anna

-----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 Amy Wade
Sent: 16 February 2009 13:33
To: r-sig-mixed-models at r-project.org
Subject: [R-sig-ME] No estimated scale value given

Hello,



I'm trying to run generalized linear mixed models using the lmer() function in the lme4 package. The problem is that the output does not give me a value for the Estimated scale. The rest of the output is as it should be. This makes it very difficult to assess whether the model is overdispersed. My colleague tried the same code on his computer and it did give an estimated scale value. I tried unistalling R and reinstalling the latest edition
(2.8.1) with the latest lme4, but this made no difference. I also tried 'summary(model)@sigma' which people suggested on the CRAN forums to extract the scale parameter. This always gave me an answer of '1' even when I used data where I knew this was not the case. When I load up the lme4 library it does warn me about several objects being masked from 'stats' and 'base'.



Any idea why the output is not giving me the estimate scale? Or any idea how to extract it somehow?



Many thanks,

Amy





Amy Wade

Centre for Agri-Environmental Research

University of Reading

Tel: +44 (0)118 987 5467

Mobile: +44 (0)7786 557 007

Email: a.s.wade at reading.ac.uk

http://www.reading.ac.uk/caer/student_amy_wade.html




        [[alternative HTML version deleted]]

_______________________________________________
R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models


The University of Aberdeen is a charity registered in Scotland, No SC013683.