--
*** Three hierarchical modeling email lists ***
(1) unmarked (this list): for questions specific to the R package unmarked
(2) SCR: for design and Bayesian or non-bayesian analysis of spatial capture-recapture
(3) HMecology: for everything else, especially material covered in the books by Royle & Dorazio (2008), Kéry & Schaub (2012), Kéry & Royle (2016, 2021) and Schaub & Kéry (2022)
---
You received this message because you are subscribed to the Google Groups "unmarked" group.
To unsubscribe from this group and stop receiving emails from it, send an email to unmarked+u...@googlegroups.com.
To view this discussion on the web visit https://link.edgepilot.com/s/8d4394fa/qqCLfdV1rkGI1TSayIHb6Q?u=https://groups.google.com/d/msgid/unmarked/ZR0P278MB086914F2E5F946C0D38815A5EB602%2540ZR0P278MB0869.CHEP278.PROD.OUTLOOK.COM.
Dear Quresh,
no, I actually don't say anything about whether it's a good or a bad idea to fit a model with random site effects in the detection part of the model. In fact, there's the result that in a cap-recap model, any unmodelled 'individual' heterogeneity will lead to the underestimation of population size. Translated to an occupancy model, this means that we will then underestimate occupancy (see Royle, Biometrics, 2006 and Dorazio, Ecology, 2007). In my post I simply deal with the question of whether we can estimate such an 'overdispersion' parameter. We can in case of the detection model, but we cannot in the case of the occupancy part of the model.
Your points in your earlier mail may well be true, but my intuition is not strong enough to say something about it. Hence, if I wanted to learn, I'd investigate by another simulation, where for instance I would simulate data with variable degrees of overdispersion in both the occupancy and the detection parts of the model and then fit a model with random site effects only in detection and compare truth and estimates.
Best regards --- Marc
From: unma...@googlegroups.com <unma...@googlegroups.com> on behalf of Quresh Latif <quresh...@birdconservancy.org>
Sent: Monday, September 16, 2024 15:29
To: unma...@googlegroups.com <unma...@googlegroups.com>
Subject: Re: [unmarked] identifiability of random site effects in static occupancy modelAh, ignore my earlier post. I see now that you were doing it to show that it is a bad idea. :)
Quresh S. Latif
BiometricianBird Conservancy of the Rockies230 Cherry St., Ste. 150, Fort Collins, CO 80521970-482-1707 (ext. 15)Connecting people, birds and land
--
*** Three hierarchical modeling email lists ***
(1) unmarked (this list): for questions specific to the R package unmarked
(2) SCR: for design and Bayesian or non-bayesian analysis of spatial capture-recapture
(3) HMecology: for everything else, especially material covered in the books by Royle & Dorazio (2008), Kéry & Schaub (2012), Kéry & Royle (2016, 2021) and Schaub & Kéry (2022)
---
You received this message because you are subscribed to the Google Groups "unmarked" group.
To unsubscribe from this group and stop receiving emails from it, send an email to unmarked+u...@googlegroups.com.