Why is there a difference between my R estimates and my Distance estimates?

25 views
Skip to first unread message

Ebony James

unread,
Nov 11, 2021, 1:49:20 AM11/11/21
to distance-sampling
Hi everyone,

First time user of Distance here. First questions I have are related to standard CDS.

Through R, I would get density estimates of 1.65/ha of an arboreal species, but in distance, it seems be to lower, 1.43/ha.

Is this a processing difference that occurs through the Distance package in R or is this an entry issue?

Kind regards,
Ebony

Res. Ass. Uğur ÖZSANDIKÇI

unread,
Nov 11, 2021, 2:50:05 AM11/11/21
to distance-sampling
Hi Ebony,
This question has been asked several times (including me :) ) before and answered by Eric. The short answer is that the likelihood values change depending on which fitting engine is used. You can find the topic from the link below
Best,

--Uğur

11 Kasım 2021 Perşembe tarihinde saat 09:49:20 UTC+3 itibarıyla elja...@outlook.com şunları yazdı:

Eric Rexstad

unread,
Nov 11, 2021, 2:54:22 AM11/11/21
to ozsandikciugur, distance-sampling
Thanks Uğur.  That is true.  Ebony, please check that you have fitted the same model in both pieces of software and also check that the same data (and truncation distances) are used in the analysis.

If you are still in doubt, contact me off list and we could look more deeply into your situation.


From: distance...@googlegroups.com <distance...@googlegroups.com> on behalf of Res. Ass. Uğur ÖZSANDIKÇI <ozsandi...@gmail.com>
Sent: 11 November 2021 07:50
To: distance-sampling <distance...@googlegroups.com>
Subject: {Suspected Spam} [distance-sampling] Re: Why is there a difference between my R estimates and my Distance estimates?
 
--
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/7e0e9ee8-beea-47b9-82e4-bca36bc64e1an%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages