RMark and fixing Psi parameters (multistrata model)

[jlbrown]

Hello, all. I'm still learning my way around RMark, and have been stymied by the appropriate formulation of Psi in a multistrata model.

In specific, I have encounter histories of tropicbirds, in which some were juveniles and the others adults. I'm trying to set up a multistate model in which juveniles can transition to adult state, but logically, not vice-versa. In standard MARK, it's quite straightforward to specify a parameter for Psi Adult-Juvenile and force it to be 0, but I've yet to succeed in doing this in RMark. I suspect the answer involves the formula specification for Psi, but nothing seems to be working when I try to use the "fixed" term using an index specification for the real parameters. My efforts are complicated by the fact that this is a very long term data set (39 encounter periods spanning 27.5 years)!

Could anyone suggest a way to accomplish this? Thanks in advance for any help, including more insight on specifying Psi formulations in RMark! The help file is very helpful to a point, but for dense folks like myself, more examples would be extremely useful.

[jlaake]

One of the current problems with the documentation is the lack of more complete examples for models other than CJS. If you have good examples that you want to share with others that would be very helpful. The examples could exclude the data or use a data subset. It is fairly time-consuming and at this point I don't have the time to do lots of examples.

I've been trying to rectify this partially by responding to queries on this list server and to individual requests. However I haven't been able to put all that material into the documentation yet. If you have a question about RMark after reading the documentation, please do a search on this list server and see if it has already been answered. I think your question was addressed Oct 2006 in a discussion with arpat. There is also a submission that I included about mlogits used with the mstrata model. If that material doesn't help send me a message (jeff.laake@noaa.gov) and I'll be glad to help and after we resolve it we can post a message for others or include an example in the documentation,

--jeff

[jlbrown]

Aha! Thanks for the tip - I had scrolled through a number of pages of posts, but had not searched far enough back in the archives. Thanks for pointing out my error rather than just ignoring my post.

It seems like the arpat thread might indeed be just what I need, and that I was not too far off with my naive coding attempts. I will go back to the virtual drawing board and try it again, and correspond with Jeff directly if further questions come up. Assuming I get things to eventually work, I'll be sure to post the appropriate code for all to enjoy.