I think it would help to deal with prediction early on - that way you can explain when AIC is optimal, and thus its irrelevance to a lot of ecological questions. You can also point readers interested in predictions to the Brewer et al. MEE paper, and then forget about prediction.
I think it would help to deal with prediction early on - that way you can explain when AIC is optimal, and thus its irrelevance to a lot of ecological questions. You can also point readers interested in predictions to the Brewer et al. MEE paper, and then forget about prediction.