Closed teixeirak closed 4 years ago
Note that this is a bit clunky-- it would be helpful to have a term for the models that were statistically indistinguishable ($\Delta$AICc<2) from the top model. We could revert to calling these top models, but "very top" sounds funny...
We could say "best models" for dAIC=0, and top models for dAIC<2? Technically models with dAIC=0 are the "best" models
Good idea; let’s go with that.
We're currently using "top models" to refer to both the very top models, and to any within dAICc =2. This needs to be checked/ fixed throughout. As I write the results, I'm using "top" to refer only to the very top.