Model averaging and psi prediction at t+1

25 views
Skip to first unread message

Yaelle Blais

unread,
May 30, 2024, 7:32:05 AMMay 30
to unmarked
Hello everybody, 

I am using a model averaging in dynamic occuancy models. I used the modavgPred function to get the prediction of psi for the first year. 

I am struggling to get t+1 predictions for psi parameter. 
I can't find the equivalent of smoothed() and @projected when using a model averaging/list of models. 

Do you have an idea ? 

Best, 

Yaelle

Marc J. Mazerolle

unread,
May 30, 2024, 8:31:24 AMMay 30
to unma...@googlegroups.com
Bonjour Yaelle,

modavgPred( ) will work to get predictions of the parameters in your
model, but not derived parameters. In the case of the dynamic occupancy
model fit with colext( ), the parameters are the occupancy of the first
season, colonization between seasons, extinction between seasons, and
detection probability on each visit. If you want to get model-averaged
predictions for derived parameters (e.g., occupancy in t+1), then you
will need to get the estimates (either by hand or smoothed) from each
model, and then plug them in the modavgCustom( ) function. See
?modavgCustom for a few examples of using the function.

Best,

Marc
--
____________________________________
Marc J. Mazerolle
Professeur agrégé et directeur du bac. en environnements naturels et
aménagés
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

-------- Message initial --------
De: Yaelle Blais <y.bl...@gmail.com>
Répondre à: unma...@googlegroups.com
À: unmarked <unma...@googlegroups.com>
Objet: [unmarked] Model averaging and psi prediction at t+1
Date: Thu, 30 May 2024 02:28:37 -0700

Yaelle Blais

unread,
Jun 17, 2024, 8:31:03 AMJun 17
to unmarked
Hello Marc,
Sorry for my late reply
Thanks a lot for your answer thats helped me in my work.
All the best, 
Yaelle

Reply all
Reply to author
Forward
0 new messages