Model averaging the derived parameters

[kanchan1234]

Hi all

I am analyzing the seasonal changes in occupancy of herbivores in my study areas using Hines et al 2014 models. I am summarizing the results and i was wondering if we could do the model averaging for the derived parameters (like derived occupancy estimates; lamda estimates)...Any suggestion in this regard is highly appreciated...

Thanks

Kanchan

[darryl]

Yes you can

[kanchan1234]

Thank Darryl... Does this follow normal model averaging procedure ?

Kanchan

[darryl]

Yes.

[Aprichard]

I am also trying to calculate model-average occupancy estimates but I am stuck at what to put in for the overall SE for a model. I think I can use the SE for occupancy of the sample (not the population), but I couldn't find a clear explanation of how to calculate it in the forum.

[darryl]

What do you mean by

overall SE for a model.

?

[Aprichard]

Please correct me if I am thinking about this wrong, but I am trying to find a model averaged occupancy for my sample of sites and a SE for that estimate. I assumed, I would take the mean occupancy and SE for each of the individual top models and then conduct model-averaging on the mean and SE's.

Is there a more appropriate way to do this, or better summary statistics to report?