[R-sig-ME] GLMM parameter interpretation

[ONKELINX, Thierry]

A quick answer: No

ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Research Institute for Nature and Forest
team Biometrie & Kwaliteitszorg / team Biometrics & Quality Assurance
Kliniekstraat 25
1070 Anderlecht
Belgium
+ 32 2 525 02 51
+ 32 54 43 61 85
Thierry.Onkelinx at inbo.be
www.inbo.be

To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of.
~ Sir Ronald Aylmer Fisher

The plural of anecdote is not data.
~ Roger Brinner

The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data.
~ John Tukey


-----Oorspronkelijk bericht-----
Van: r-sig-mixed-models-bounces at r-project.org [mailto:r-sig-mixed-models-bounces at r-project.org] Namens Adam Smith
Verzonden: vrijdag 20 januari 2012 5:22
Aan: r-sig-mixed-models at r-project.org
Onderwerp: [R-sig-ME] GLMM parameter interpretation


All,

A quick question, and an easy one I expect.

Suppose the following generic overdispersed log-linear model:

Count ~ A + B + C + (1|level1) + (1|level2) + (1|obs)

A, B, and C are all factors.
A does not vary within level1 subjects.
B does not vary within level2 subjects.
C varies within level 1 and level 2 subjects.

Does the conditional nature of the GLMM and the lack of variation in A (within level 1 subjects) and B (within level 2 subjects) result in nonsensical or uninterpretable coefficients for these two covariates?

Thanks,

Adam Smith
Dept. Natural Resources Science
105 Coastal Institute in Kingston
University of Rhode Island
 		 	   		  
_______________________________________________
R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models