[R-sig-ME] Problems with warnings

[Mollie Brooks]

Your attachments were removed by the system, so I can’t see the code. 

It sounds like you might need to add `+0` in your random effect specification.
Have you looked at the AR(1) documentation in this vignette?
https://cran.r-project.org/web/packages/glmmTMB/vignettes/covstruct.html <https://cran.r-project.org/web/packages/glmmTMB/vignettes/covstruct.html>
There it shows the term ar1(times + 0 | group)

cheers,
Mollie

> On 20 Jan 2022, at 10.29, Mathias Emil Kaae <mek using ecos.au.dk> wrote:
> 
> To whom it concerns,
> 
> I am using the package glmmTMB to fit a mixed model with interactions using the beta-binomial distribution. I am trying to model plant cover data for specific species in a habitat using a randomized block design with five blocks and 16 plots within them. The question is: are this plant species increasing/decreasing its cover within different plots as a result of different treatments?
> 
> As I see it, I only get non-significant results (which is fine). However, I do get this warning when using the package glmmTMB:
> 
> Warning message:
> In getReStruc(reTrms, ss, aa, reXterms, fr) :
> AR1 not meaningful with intercept
> 
> Should I worry, and how can I solve this?
> 
> I have attached the code and a subset of the data.
> 
> Mathias Emil Kaae
> PhD student, Aarhus University
> Institut for Ecoscience
> Vejlsøvej 25, 8600 Silkeborg
> Mobil nr.: + 45 61 72 14 01
> E-mail: mek using bios.au.dk<mailto:mek using bios.au.dk>
> 
> 
> _______________________________________________
> R-sig-mixed-models using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models


	[[alternative HTML version deleted]]