GLMM power analysis
This is typically done by simulation. Googling "glmm power
simulation" gives a lot of useful hits. There are R packages:
library(sos)
sos::findFn("{power analysis} mixed simulation")
turns up the fullfact, pamm, simr, simglm packages ... or you can write
it yourself (the simulate() method.
There are a variety of publications on this subject, too:
Johnson, Paul C. D., Sarah J. E. Barry, Heather M. Ferguson, and Pie
M?ller. ?Power Analysis for Generalized Linear Mixed Models in Ecology
and Evolution.? Methods in Ecology and Evolution 6, no. 2 (February 1,
2015): 133?42. https://doi.org/10.1111/2041-210X.12306.
Kain, Morgan P., Ben M. Bolker, and Michael W. McCoy. ?A Practical Guide
and Power Analysis for GLMMs: Detecting among Treatment Variation in
Random Effects.? PeerJ 3 (September 17, 2015): e1226.
https://doi.org/10.7717/peerj.1226.
working on updating the GLMM FAQ ...
On 18-03-20 10:24 AM, Louise Heitzmann wrote:
Hi, I am working on seals' behaviour related to 6 treatments (2*3 treatments tested seperately over time). My response variable is a counting (number of nasal openings) which follow a Poisson Distribution. The experimental design is : 2 seals' individuals in a box, there is 7 box, and experiments are repeated during 7 days. Each pair of individuals are tested in their specific box, but not all pairs are tested every day : sometimes only 5 pairs . Thus, I have between 6 and 10 repeated mesures per individual per treatment and I have a nested (Box,Individuals) and crossed (Days) design. My model looks like : y = ?0 + ? * Treatment + ( 1|Box/individus) +(1|Days) I want to do a power analysis to test the ability to detect a true effect. Thus, I was wondering how can I make a power analysis related to my model ? Thank you for considering my request, Louise heitzmann