Is there no cross validation package that works for anova/aov objects? I can't find anything useful. Want to compute cv eta squared. #rstats
-
Any reason for using aov object instead of equivalent lm object (for which there are many tools)?
Yes, I want eta^2 for the categorical predictors as a whole, not betas for each level.
-
-
Never seen this for individual predictors or groups thereof. Could imagine a definition, but surprised if any packages do it
-
It's useful bc large betas for some levels do not mean categorical predictor has large effect. Depends on base rates of levels.
-
Multiple packages provide eta^2 (lsr, ez, heplots, ...), but only the regular version, not cross validated.
-
But I guess I will have to expand my cross validation function to cover this use case since it seems to be missing.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.