Defining capture and recapture prob. in Robust Design

[ant]

I'm using the RD (the simplest closed option) to estimate N and get estimates of survival.

My doubt concerns of how should I set p and c:

1- perform a closed analysis (perhaps I Huggins, because I wouldn't be interested in N), in each of my 13 periods of five sampling days to view how p and c goes in each primary occasion?

2- choose the simplest way, setting p=c.

3- Other option!

Thank you for helps in this post, as well in many others that have already helped me,
Antonio

[sbonner]

Hi Antonio,

My first question is to ask whether or not you have any reason to believe that there is a difference between the initial capture and recapture rates within each secondary period. If there is, then you need to be fitting models that allow for p and c to differ.

If you have no reason to believe that p and c should differ, then I would recommend that you fit all of your models with p=c. As a check you might refit the best model(s) allowing for p and c to vary. If you find that the new model(s) fit better, then it's time to look at the actual estimates to see where the differences are and whether or not they can be explained.

Fitting separate models for each secondary period is double dipping and though I don't think it would be a big problem in this case, I don't see a real need for it. It should be simple enough to study this structure within the RD itself.

Anyone else have different thoughts?

Simon

[ant]

Well, thank you very much Simon!

You did make the things clear for !