AIC values for IP SECR

questions concerning anlysis/theory using program DENSITY and R package secr. Focus on spatially-explicit analysis.

AIC values for IP SECR

Postby sam.meyler » Thu Oct 21, 2010 11:24 am

Hi,
I cannot find AIC values for different models run through IP SECR. Perhaps confusion is because Efford 2004 does not seem to choose his models through AIC, yet the help section in Density shows AIC on the output format of IP SECR... Is it that this is a newer version? How does one choose the most appropriate closed captures model without AIC in output?
Thanks again for your time! :?
Sam
sam.meyler
 
Posts: 2
Joined: Thu Oct 21, 2010 8:02 am

Re: AIC values for IP SECR

Postby murray.efford » Thu Oct 21, 2010 5:17 pm

Sorry I missed this earlier query somehow.

IP SECR does not use maximum likelihood and therefore AIC cannot be used for comparing models (there really aren't many model options in any case). The AIC that appears on the output lines in the rather old help example (version 3.2, 2004!) relates to the closed-population N-hat, now called CPNhat.

This is one of several reasons to prefer ML SECR. IP SECR still has a narrow role in allowing the unbiased estimation of g0 for data from single-catch traps when traps are saturated (see Efford et al 2009 in the Thomson, Cooch & Conroy book). I suspect that is not an issue for mouse lemurs.

Murray
murray.efford
 
Posts: 686
Joined: Mon Sep 29, 2008 7:11 pm
Location: Dunedin, New Zealand


Return to analysis help

Who is online

Users browsing this forum: No registered users and 14 guests