[R-sig-ME] Covariance matrices in GLMER

[Voeten, C.C.]

Dear Diana,

1. No, both lmer and glmer by default use unstructured covariance matrices, although they can be massaged into using diagonal, see afex::lmer_alt. Those are the only two options.
2a. Support for other covariance matrices is available in MASS::glmmPQL (which I would not recommend; it inherits all the limitations of nlme::lme and uses PQL) and in glmmTMB::glmmTMB. Depending on what exactly you need to do, glmmTMB is probably what you're looking for.
2b. For glmer and glmmTMB, the ranef function will probably give you what you need.

Best,
Cesko

> -----Oorspronkelijk bericht-----
> Van: R-sig-mixed-models [mailto:r-sig-mixed-models-bounces using r-
> project.org] Namens Kornbrot, Diana
> Verzonden: dinsdag 30 oktober 2018 10:16
> Aan: r-sig-mixed-models using r-project.org
> Onderwerp: [R-sig-ME] Covariance matrices in GLMER
> 
> It seems to me that GLMER always uses diagonal matrices.
> 1. Is this correct?
> 2. If not, (a) how does one specify different covariance matrices (b) how
> does one get output of covariance matrices that GLMER actually used/
> Thanks for all help Diana _____________________________________
> Professor Diana Kornbrot
> Mobile
> +44 (0) 7403 18 16 12
> Work
> University of Hertfordshire
> College Lane, Hatfield, Hertfordshire AL10 9AB, UK
> +44 (0) 170 728 4626
> d.e.kornbrot using herts.ac.uk<mailto:d.e.kornbrot using herts.ac.uk>
> http://dianakornbrot.wordpress.com/
> http://go.herts.ac.uk/Diana_Kornbrot
> skype:  kornbrotme
> Home
> 19 Elmhurst Avenue
> London N2 0LT, UK
> +44 (0) 208 444 2081
>  ------------------------------------------------------------
> 
> 
> 
> 
> 	[[alternative HTML version deleted]]
> 
> _______________________________________________
> R-sig-mixed-models using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models