Model Averaging

[Wijid]

Hi, I have another question. I am not sure if there is a set method for doing this, as far as I can tell it isn't in the handbook. If, for example, I have two models in my candidate set (trying to keep it simple):
Phi(.)p(g)
Phi(.)p(g*t)
I am interested in estimating the recapture rate for each group in the population (g), and the group model comes out as the best model.

I know model averaging is the better method for estimating parameters. If I model average over these two models though I will get recapture rates for each time period as well as each group because of the second model. So then can I average the recapture rate values for each group (across the time periods)?

Is this a viable method or am I better to ignore model averaging and just take the estimates from the best model, in this case the group model?

Thank you for your time