Categorical interaction and crashing covariate plot

[mhudson]

Hi all,

First off I just want to say thanks to all of you - I've been surfing this forum for a while and it's been tremendously helpful! One suggestion: if you've solved your own problem, please consider posting your solution since many others might encounter the same problem. You'll save yourselves a bunch of emails wondering at the solution.

Now for my 2 questions. I'm using the nest survival model in Program MARK, and I've modeled a mix of continuous and categorical variables.

One of my questions of interest is how nest substrate and its arrangement (both categorical/indicator variables coded using dummy coding) interact and what the specific DSRs are for each level of interaction.

I have 3 levels of substrate interacting with 2 levels of arrangement, e.g., S1A1, S1A2, S2A1, etc. My question is this: is there any way to figure out what the DSRs for the reference cells are, since the reference cells from both variables will produce a zero?

And the second question: does anyone have a problem with the individual covariate plot function when using the nest survival model? Every time I try to plot a single covariate, a very long error message pops up (and is rarely the same), forcing me to "abort". This happens whether I'm trying to plot a single variable's effect on DSR or whether it's a more complex model.

Thanks so much, and sorry for the long post.
Marie-Anne

[geoffwah]

Hi Marie-Anne and All

Just throwing my hat into the ring re the individual covariate plot function. It's great when it works - really useful. However, I too have had many occasions where it 'aborts' the program (I'm running occupancy models), or produces some pretty crazy 95%CI's....

Cheers

Geoff