Is it possible to fit occupancy models with brms?
Occupancy models are used in ecology and estimates the probability a site being occupied by a species (ψ) and the probability of detection of the species §. From my basic knowledge I think they are zero-inflated binomial models.
Best,
Manuel
I don’t know about occupancy models, but you can fit zero-inflated binomial models via family zero_inflated_binomial.
Thank you very much Paul.
I know Bob Carpenter and Richard Erickson has been working on occupancy models on Stan and Rstan but I have not seen an example on the Stan forum of the implementation in brms, which will be very helpful and interesting. I started to understand and works on bayesian statistics because brms and rstanarm.
Manuel
I have written a set of tutorials on doing Occupancy Modeling with Stan, which is on GitHub (linked here). These are a work in progress, although l will likely move them to my package (see next paragraph).
I am also writing a package, rstanOcc, which is rough shape for doing occupancy models with rstan. The package is on GitHub (linked here). I am able to install the package on my machine, but my collaborators current are having trouble.
Also, as a tip for Discourse, if you tag people (e.g., @Richard_Erickson), we can more easily see what you’re doing.
Edit: Oct 2024, here’s the new link UMESC / quant-ecology / Occstanhm · GitLab (usgs.gov)