Covariate help

questions concerning analysis/theory using program PRESENCE

Covariate help

Postby not_present » Wed Sep 21, 2016 12:35 pm

Hi-

I am running a single-season, two species model. I have one survey covariate. I insert a column and use the covariate name in the last column of the detection matrix. When I run the model, the output provides an estimate for the covariate for species A (pA[1].covariate) in the results table. I need the estimate for pB also. What am I doing wrong? Thanks.
not_present
 
Posts: 10
Joined: Fri Apr 08, 2016 1:31 am

Re: Covariate help

Postby jhines » Wed Sep 21, 2016 2:19 pm

If you added 1 column to the detection design matrix, then only 1 beta estimate will be produced by PRESENCE. That beta value will be labelled with the first "real" parameter name found in the design matrix which has a non-zero value for that column (pA in your case). If that covariate is also in the rows for pB, then the effect for pB is the same as for pA. If the covariate is in all rows of the detection design matrix, then the effect will also apply to rA, rBA and rBa, though you won't see a separate beta for the other "real" parameters.

If you had another column in the design matrix with the covariate in the rows for pB only, then you would get a beta estimate for pB. This model would be different from the one you mentioned in that there would be a different effect of the covariate on detection of species A than species B.
jhines
 
Posts: 599
Joined: Fri May 16, 2003 9:24 am
Location: Laurel, MD, USA

Re: Covariate help

Postby not_present » Tue Oct 04, 2016 8:32 am

Thanks!
not_present
 
Posts: 10
Joined: Fri Apr 08, 2016 1:31 am


Return to analysis help

Who is online

Users browsing this forum: No registered users and 10 guests

cron