model/design assistance?

questions concerning analysis/theory using program MARK

model/design assistance?

Postby res_2 » Thu Dec 06, 2007 11:20 pm

Hi,
This is my first time using MARK and I want to make sure that I am going about it as correctly as possible. (I have read the manual, but have not been able to resolve all my questions). The goal is to get a reasonable population estimate for snakes in a geographically closed area. Here are the basic constraints and attempted model with questions following:

Mt+1 = 440
approximately 19% recaptures
continuous trapping for 3-1/2 months arbitrarily split into 14 occasions
(= 7 days in each "occasion" with no time in between “occasions”)
Huggins closed capture model
The global model with 2 groups is fully time variable and includes an individual covariate, 2 occasion covariates and all interaction terms (all covariates were standardized with the product function so that they fall between 0 and 1). The models were run with the logit link. For this model and nested models, the data is sparse and MARK is not correctly counting all theoretically estimable parameters (see examples below).
C-hat (from the most-highly parameterized model without the individual covariate) is around 6.0.

**Question 1) Is this parsing of continuous trapping (without time off between occasions) is acceptable?**

**Question 2) Is it permissible to use Huggins given the slight violation of the no birth/no death assumption and the lack of time between “occasions”?**

**Question 3) How do I count the # of estimable parameters given the above model? (From reading the manual, it is clear to me how to count parameters in a CJS model without covariates, but I am not confident about counting in the highly parameterized Huggins model).**

Here are some examples of the real and beta parameter estimates from the most parameterized model:

12:p 0.7354565 0.1622926 0.3515005 0.9344670
13:p 1.0000000 0.1261462E-06 0.9999998 1.0000002
14:p 1.0000000 0.3516278E-09 1.0000000 1.0000000
15:p 0.1482841 0.0292044 0.0996294 0.2150250
32:c 0.0465300 0.0213302 0.0186633 0.1112866
33:c 0.0364086 0.0169086 0.0144785 0.0885700
34:c 0.1388794E-10 0.0000000 0.1388794E-10 0.1388794E-10
35:c 0.0043216 0.0053880 0.3728074E-03 0.0480838

53:p g1*t10*effort -104.01053 253.22548 -600.33249 392.31142
54:p g1*t11*effort -106.64894 205.43424 -509.30006 296.00218
55:p g1*t12*effort -103.27246 401.21947 -889.66264 683.11771
56:p g1*t13*effort 365.14474 2326.4007 -4194.6006 4924.8901
57:p SVL 47.938602 0.0000000 47.938602 47.938602
58:p g1*SVL -601.62551 0.0000000 -601.62551 -601.62551

**Main question/ request!) If anyone has suggestions on how to better deal with this dataset, I’d greatly appreciate them.**

thanks.
res_2
 
Posts: 1
Joined: Thu Dec 06, 2007 1:15 pm

Return to analysis help

Who is online

Users browsing this forum: No registered users and 3 guests

cron