Hi,
I am calculating the relative variable importance value of different predictors using AIC model selection procedures.
It seems logical that " that if I include interactions, then I can only compare the Relative Importance Weights (RIWs) of main effects with each other, and I can only compare the RIWs of interactions with each other. I cannot compare main effect RIWs with interaction RIWs (because main effects are present in more models than interactions)."
but, anyone can provide me a citation for this?
thanks