Inference from models with unidentifiable parameters POPAN

questions concerning analysis/theory using program MARK

Inference from models with unidentifiable parameters POPAN

Postby butterphi » Wed Feb 09, 2022 10:48 am

I am interested in daily population sizes and the gross population size of a butterfly population.
To adress my question, I use a POPAN JS-model.
As far as I understood the MARK book, it is not recommended to make pent constant, even more so because it would not be biologically reasonable in my system. What I should actually do, is to model entrance depending on time (and sex). However, adding this complexity to my model comes with the problem of 'extrinsic' unidentifiability for some of the pents.

I read various related posts in the forum but it is still unclear to me how to deal with estimates from a model with a few unidentifiable pent-parameters. Are the other parameter estimates nevertheless valid? The pent itself is not of interest to me but I suspect that the other parameter estimates may be affected? Nevertheless, going back to a less realistic model (i.e. pent(.)) just because of two or three unidentifiable pents would be quite a pity...

Many thanks for your advice!
butterphi
 
Posts: 9
Joined: Tue Nov 02, 2021 9:00 am

Return to analysis help

Who is online

Users browsing this forum: No registered users and 11 guests