Mark-Resight garbage estimates

questions concerning analysis/theory using program MARK

Mark-Resight garbage estimates

Postby Harold » Mon Feb 19, 2018 4:41 pm

Hello,

I am trying to estimate a Poisson Mark-Resight Model. However, I am getting obviously garbage U estimates.

Following the recommendations in the Book, I tried a range of initial values and options on the "run" window. However, I continue to get similar garbage estimates. I also tried fixing sigma to a value from a similar but different experiment, and got a credible U estimate, but a much higher deviance.

Anything else I can try?
--
Or perhaps my data is insufficient...
My data: 100 individually-identifiable marked individuals, 27 of which were resighted at least once (and up to 22 times) during the single resight period. 367 unmarked individuals were observed.

Thanks!
Harold
Harold
 
Posts: 2
Joined: Mon Jan 15, 2018 5:30 pm

Re: Mark-Resight garbage estimates

Postby bmcclintock » Wed Feb 21, 2018 2:12 pm

Hi Harold,

Sounds like a convergence issue to me, but it's difficult to diagnose based on the information provided. Have you tried fixing sigma=0 and using estimates from this model as the initial values?

Cheers,
Brett
bmcclintock
 
Posts: 44
Joined: Mon Feb 12, 2007 6:10 pm
Location: NOAA National Marine Mammal Laboratory

Re: Mark-Resight garbage estimates

Postby Harold » Sat Feb 24, 2018 6:27 pm

Hi Brett,
Thanks for your help!
Have you tried fixing sigma=0 and using estimates from this model as the initial values?

I just tried this, buts still get garbage estimates.
Is there any other diagnostic info I can provide?

Thanks!
Harold
Harold
 
Posts: 2
Joined: Mon Jan 15, 2018 5:30 pm


Return to analysis help

Who is online

Users browsing this forum: Google [Bot] and 2 guests

cron