[R-meta] Model selection using glmulti & MuMln with multilevel (rma.mv) models

[Viechtbauer, Wolfgang (SP)]

Hi Mia,

Since you mentioned multilevel, I take it you want to use rma.mv() then?

The example here would be easy to adapt:

http://www.metafor-project.org/doku.php/tips:model_selection_with_glmulti_and_mumin

Basically, it would just require something like:

rma.glmulti <- function(formula, data, ...)
   rma.mv(formula, vi, random = ~ 1 | study / effect, data=data, method="ML", ...)

If you are also interested in interactions, you should read the "Models with Interactions" section. The number of models quickly explodes once one also wants to consider interactions.

Best,
Wolfgang

-----Original Message-----
From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces using r-project.org] On Behalf Of Mia Daucourt
Sent: Friday, 06 March, 2020 16:37
To: r-sig-meta-analysis using r-project.org
Subject: [R-meta] Model selection using glmulti & MuMln with multilevel (rma.mv) models

Good morning,

I am interested in using the glmulti or MuMln package to choose the best combination of moderators for a multi-level model with random effects for study and effect size included. I am also interested in examining interactions between moderators. I need your help, please! Do you have any advice on how to code this or any resources you can point me to?

Please let me know!

My best,

Mia