False positive occupancy models

[joeflangley]

Hi there,

I was wondering whether there are currently any plans to add functionality for dealing with false positives in flocker, e.g like occuFP in unmarked.

Cheers,
Joe

[jsocolar]

This would be a moderate lift--nothing too difficult conceptually, but quite a lot of bookkeeping, particularly to ensure that all of the post-processing functionality works correctly. I don't have time to lead this myself, but if you or anybody wants to take a stab at it I'm very happy to help. I would estimate that for somebody experienced with R but not R package development, it would probably take on the order of days to have a working branch where single-season versions of these models can be fitted but not post-processed (and again I'm happy to help!), somewhat longer to get all of the post-processing lined up, and longer still to extend this framework to multiseason and/or data augmented models.

The first step, for anybody who is interested, is to write out carefully the following two things:

1. The model likelihood
2. A clear description of what extra data needs to be passed to the likelihood, beyond detection/non-detection data and covariates.

[joeflangley]

Thanks for this, Jacob. I might look into it in the coming months, so I may well be in touch!