Can't run Bootstrap GOF with occupancy model

questions concerning analysis/theory using program MARK

Can't run Bootstrap GOF with occupancy model

Postby sixtystrat » Wed Mar 22, 2006 10:01 am

My computer crashes when I try to do a bootstrap GOF with an occupancy model. I want to estimate expected and observed deviance so I can estimate c-hat. Everything else seems to work fine and my other data sets seem okay. I have uninstalled and reinstalled Mark several times to no avail. Thanks for any help!
Joe Clark
sixtystrat
 

RE: Can't run Bootstrap GOF with occupancy model

Postby darryl » Wed Mar 22, 2006 5:20 pm

Hi Joe

Don't quote me on this one, but I don't think the bootstrap GOF test is designed to be used with the occupancy models; perhaps that is causing the crash.

There is however a test that has been specifically developed for occupancy models that is implemented in the software PRESENCE (downloadable from http://www.mbr-pwrc.usgs.gov/software.html). It's only designed for the single season/year analysis. The actual method is described in MacKenzie and Bailey (2004) Assessing the fit of site occupancy models, JABES. You can download a pdf of this paper from my bio page at http://www.proteus.co.nz

Hope this helps
Darryl
darryl
 
Posts: 498
Joined: Thu Jun 12, 2003 3:04 pm
Location: Dunedin, New Zealand

Postby sixtystrat » Thu Mar 23, 2006 10:09 am

Thanks Darryl. You may be right. I'll get your paper and do some more reading.
Joe
sixtystrat
 

Can\'t run Bootstrap GOF with occupancy model

Postby gwhite » Thu Mar 23, 2006 7:33 pm

Joe:

I have posted a new MARK version that fixes this bug, plus also allows you to run the median c-hat procedure with simple occupancy data. You can also construct the chi-squared test mentioned by Darryl by dumping the observed and expected values into an Excel spreadsheet and computing a Pearson chi-square statistic. Also, the Pearson chi-square statistic is printed in the full output, and this value would be generally appropriate for simple occupancy data.

Gary
gwhite
 
Posts: 340
Joined: Fri May 16, 2003 9:05 am

Re: Can\'t run Bootstrap GOF with occupancy model

Postby cooch » Thu Mar 23, 2006 8:13 pm

gwhite wrote:Joe:

I have posted a new MARK version that fixes this bug, plus also allows you to run the median c-hat procedure with simple occupancy data. You can also construct the chi-squared test mentioned by Darryl by dumping the observed and expected values into an Excel spreadsheet and computing a Pearson chi-square statistic. Also, the Pearson chi-square statistic is printed in the full output, and this value would be generally appropriate for simple occupancy data.

Gary


Which has just been mirrored at

http://www.phidot.org/software/mark/download/
cooch
 
Posts: 1654
Joined: Thu May 15, 2003 4:11 pm
Location: Cornell University

Postby sixtystrat » Thu Mar 30, 2006 1:57 pm

Thanks guys. I'll check it out.
joe
sixtystrat
 


Return to analysis help

Who is online

Users browsing this forum: No registered users and 2 guests