Psi estimate as 1 in spatial correlation model

questions concerning analysis/theory using program PRESENCE

Psi estimate as 1 in spatial correlation model

Postby sprihharsh » Mon Sep 28, 2015 1:43 pm

Hi,

I am having issue while analyzing my data through spatial correlation model. I have data from occupancy survey for three carnivores tiger, leopard and sloth bear with decent signs that is 1 recorded for each species. The sample size is 43 with 8 replicates. I have 7 site covariates: Forest type entered as 00, 01 for two types; terrain type entered as 000,100,010 for three types, distance from village and distance from road both z transformed. I also have 2 survey covariates: prey presence and human disturbance which are entered in terms of 0 or 1 based on their availability on the survey site.

On analysis with psi(.) th0, th1, p(), th0pi(), I am getting everything fine but as soon as I start putting some covariates and that too again in not all combination of covariates but some I am getting psi value as 1 like this:
Individual Site estimates of <psi>
Site estimate Std.err 95% conf. interval
psi 1 Pratapgarh : 0.0000 0.0000 0.0000 - 1.0000
psi 2 Murwahi : 1.0000 -1.#IND 1.#QNB - 1.#QNB
psi 3 Chapri : 1.0000 -1.#IND 1.#QNB - 1.#QNB
psi 4 Salhewara : 1.0000 -1.#IND 1.#QNB - 1.#QNB
psi 5 Banjaari : 1.0000 -1.#IND 1.#QNB - 1.#QNB
psi 6 Mandlakona : 1.0000 -1.#IND 1.#QNB - 1.#QNB
psi 7 Bhawartonk : 1.0000 -1.#IND 1.#QNB - 1.#QNB
psi 8 Dhandabra : 1.0000 -1.#IND 1.#QNB - 1.#QNB
psi 9 Jamunpaani : 1.0000 -1.#IND 1.#QNB - 1.#QNB

Some I am getting mix of both 1 and <1 values like this:
Individual Site estimates of <psi>
Site estimate Std.err 95% conf. interval
psi 1 Pratapgarh : 0.0000 0.0000 0.0000 - 1.0000
psi 2 Murwahi : 1.0000 -1.#IND 1.#QNB - 1.#QNB
psi 3 Chapri : 1.0000 -1.#IND 1.#QNB - 1.#QNB
psi 4 Salhewara : 1.0000 0.0000 0.0000 - 1.0000
psi 5 Banjaari : 0.8620 0.1578 0.3169 - 0.9883
psi 6 Mandlakona : 1.0000 -1.#IND 1.#QNB - 1.#QNB

Why I am getting this type of result? Please help me in solving this issue.

Thanks & Regards,
Sprih
sprihharsh
 
Posts: 4
Joined: Sun Aug 24, 2014 12:43 pm

Re: Psi estimate as 1 in spatial correlation model

Postby Eurycea » Mon Nov 02, 2015 5:09 pm

There could be a few things going on here, but if you are entering 0 for an entire column of a covariate, that will obviously give you an answer of 0. So, for a binary covariate (e.g., forest, no forest) you'd enter a 1 or a 0 indicating either one in the same column.
Eurycea
 
Posts: 99
Joined: Thu Feb 25, 2010 11:21 am


Return to analysis help

Who is online

Users browsing this forum: No registered users and 12 guests