New models in flocker

[mhollanders]

Hi there,

I'm currently working on occupancy models specifically geared towards passive acoustic monitoring, where sites are allowed to chance occupancy state throughout the acoustic deployment period and the observation process deals with the waiting times. I'm currently working on the manuscript and it would be great if these models were available in existing R packages, for which I think flocker would make the most sense. Are you open to something like this?

Cheers,

Matt

[abfleishman]

@mhollanders I too am very interested in models like you describe. Giant +1!

[jsocolar]

Let's try to do this, though I cannot make any promises about timescales for getting things done from my end. If somebody is willing to do a big chunk of the lifting, I can comment to doing any necessary reviews and polishing in a timely manner.

To get started, check out the instructions here https://github.com/jsocolar/flocker/blob/main/inst/rmd/adding_a_family.Rmd

The main thing is to get a sketch of what the likelihood should look like, and then a sketch of what the flockerdata format for this model type needs to look like (if different from the format for one of the existing model types). I recommend against starting more substantial work on the code prior to agreeing on the data format. Usually, the biggest challenges involve getting the indexing right during the data packing (in R code) and the data unpacking (in Stan code), and it's possible to waste a lot of time here by racing ahead before we agree to the data format.