Open JustinCally opened 2 years ago
Hi Justin,
Thanks, this is really cool! I have to admit I don't know that much about calculating a Bayes R2 for this type of model so I will have to look into it (thanks for including the link). It does seem like something that would be useful to add. Are you aware of any papers that calculate a Bayes r2 for occupancy models?
Ken
Hi Ken,
I'm unaware of any paper's calculating Bayes R2 for occupancy models. Hence I'm a bit unsure whether the function written above is the preferred approach. In comparison, the method used in unmarked::modSel()
that calculates R2 based on the likelihoods of the null model and fitted model (Nagelkerke 1991). I wonder whether it is possible to adapt this method in a Bayes context...
I suppose one good thing about the function above is that it provides the bayes R2 for both detection and state submodels, which may be beneficial in understanding the value of using an occupancy model for a given situation.
Hi Ken,
I'm not sure whether the use of a bayes R2 would be broadly suitable for occupancy models given the limitations with its application in logistic regression and often low values. However I had a crack at getting a bayes R2 for the state, detection submodels (and their combined value) using the residual method (https://avehtari.github.io/bayes_R2/bayes_R2.html). A bit of a messy function and example below.
Created on 2021-11-11 by the reprex package (v2.0.1)