Using c-hat value to adjust CI

[Hannah Redmond]

Hi All,

I really appreciate this group as previous posts have helped me answer many questions.

I am using the colext function to run global dynamic occupancy models for several bird species. After running a mb.gof.test I see that, understandably, these models are "overdispersed". Most global models have a c-hat value ~ 2 but I have some that are ~4 and 10+.

I am trying to find a way to inflate the confidence intervals for my beta estimates, but so far have only seen c-hat incorporated into model predictions (modavgPred).

I am hoping for two things:

1. Beta estimates with c-hat inflated confidence intervals 

2. Partial response values with c-hat inflated confidence intervals. I have been using predict() to get partial-response values up until this point.

As for my models with c-hat > 4, should I avoid any inference on these models altogether?

Thank you in advance to anyone who has insight,

Hannah

[Marc J. Mazerolle]

Hi Hannah,

modavgPred( ) will produce CI's adjusted for overdispersion if you use
the c.hat argument. So will the functions modavgEffect( ), modavg( ),
and modavgShrink( ).

If you want CI's adjusted for c-hat for beta estimates of a single
model, you can also use the summaryOD(your.model, c.hat).

You will find examples for each on their respective help pages.

Best,

Marc
--
Marc J. Mazerolle
Professeur agrégé
Département des sciences du bois et de la forêt
2405 rue de la Terrasse
Université Laval
Québec, Québec G1V 0A6, Canada
Tel: (418) 656-2131 ext. 407120
Email: marc.ma...@sbf.ulaval.ca

Veuillez noter que je suis en télétravail. Le meilleur moyen de me
rejoindre est par courriel.

-------- Message initial --------
De: Hannah Redmond <hred...@udel.edu>
Répondre à: unma...@googlegroups.com
À: unmarked <unma...@googlegroups.com>
Objet: [unmarked] Using c-hat value to adjust CI
Date: Mon, 03 Oct 2022 11:33:51 -0700