Parameter names in PRESENCE and online book

[gferraz]

Hi, Can anyone please confirm if 'psi' and 'R' in the PRESENCE output for single-season multi-state models are the same as 'psi1' and 'psi2', respectively, in chapter 16 of the Donovan and Hines online book?
Thank you!

[jhines]

Hi Gonçalo,

PRESENCE is using the psi, R parameterization in the single-season model, not the psi1,psi2 parameterization.

Jim

[gferraz]

Hi Jim, Thanks. Where can I find a description of the psi - R parameterization? I looked in the book and in Nichols et al 2007 (Ecology) but couldn't find one.
G.

[jhines]

The multi-season multi-state model is defined in the PRESENCE help file in terms of psi,R and gives the relationship between the two. Basically, the difference is psi1,psi2 are unconditional occupancy probabilities, and in the other parameterization, psi is the unconditional probability of occupancy and R is the probability of being in state2 (eg. breeding), given that the site is occupied. So,

psi1=psi
psi2=psi*R

The advantage of the psi,R parameterization is that the two parameters only have to be constrained to be between zero and one. The psi1,psi2 parameterization needs the two parameters to be constrained to be between zero and one, and the sum of the two parameters must be less than 1.