www.phidot.org

I am using RMark to do a robust design analysis on a long time series (10 years) in 3 sites with very small populations. I would like to combine all the data in one analysis with site as a group. The problem is that the number of secondary series varies per site per year (the species' ecology drives this). Is there a way to deal with this in RMark?

For example, in separate by site runs, the time.interval parameters might be:

site 1:

Code: Select all

time.interval = c(0,1,0,0,1,0,0,0,0,0,0,1,0,0,0,0,1,0,0,0,0,0,0,0,0,0,1,0,0,0,0,1,0,1,0,0,0,0,1,0,1,0,0,0,0)


site 2:

Code: Select all

time.interval = c(0,0,1,0,0,0,0,1,0,1,0,1,0,0,0,1,0,0,0,0,0,1,0,0,0,0,1,0,1,0,0,0,0,1,0,0,0)

site 3:

Code: Select all

time.interval = c(0,0,0,1,0,0,0,0,1,0,1,0,0,0,0,1,0,1,0,1,0,0,0,0,1,0,0,0,0,0,0,0,0,1,0,0,0,0,1,0,0)

I don't know if this makes any sense, but I did a trial filling out the capture history for 1 site with . (missing) to have standard secondary intervals. I got similar but not exactly the same results.

Any suggestions? All help would be much appreciated!

Did you fix p to 0 for the padded occasions specific to the groups using fix in the design data?

Thanks for the help.
No I didn't think of that and will try it.
So this approach should work if I get the model structure right?

Also, I just posted a question on how to fix parameters for another purpose.
I am a novice in Rmark, but will get there....
Thanks much!

Yes it should.

Fixing p's worked. I got the exact same results as the 'non-padded' version.
Thanks much!