MCDS in Program Distance vs. R

[abrun...@gmail.com]

Hi All,

I apologize for the simplistic question. I cannot find a direct answer to it. I am using MCDS to get density estimates by geographic stratum where there are not enough detections per stratum to use CDS fully stratified model. My analysis is separated and conducted separately by strata.  In R, I understand using the command "formula =~Region.Label"  to accomplish this. 

I also understand that in Program Distance,  I select the MCDS analysis engine, check "use layer type stratum" and tick off both the global and stratum boxes in the detection function row. My question is, to replicate what I did in R, do I also need to add stratum as a covariate?  Or is ticking off both those boxes enough? 

Thank you, 

Andrea

[Eric Rexstad]

Andrea

Ticking both boxes in DistWin will not fit a model that uses stratum as a covariate in the detection function. So you do need to specify a model formula in DistWin as well as ticking of both boxes to duplicate the analysis you describe in R.

---

From: distance...@googlegroups.com <distance...@googlegroups.com> on behalf of abrun...@gmail.com <abrun...@gmail.com>
Sent: 15 May 2024 20:39
To: distance-sampling <distance...@googlegroups.com>
Subject: [distance-sampling] MCDS in Program Distance vs. R

 

--
You received this message because you are subscribed to the Google Groups "distance-sampling" group.
To unsubscribe from this group and stop receiving emails from it, send an email to distance-sampl...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/distance-sampling/41a0482d-f289-4ea7-8792-ac2bce1ed36cn%40googlegroups.com.