@StackStats if levels not too high, use stratified sampling in CV partitioning?
Any of you know a good solution for this problem with cross-validation? @lakens @randal_olson @gwern http://stats.stackexchange.com/questions/213837/k-fold-cross-validation-nominal-predictor-level-appears-in-the-test-data-but-no … #rstats
-
-
-
Another idea, yes.
End of conversation
New conversation -
-
-
Manually unfold the categorical into dummy variables? Probably causes singularities or something...
-
Wouldn't be a problem if you used Bayesian regression since without examples just samples from priors.
-
(Binary dummies mean that the lm always 'sees' each level but may always be 0s in the training subset)
-
Not sure how you would do this. The level is still there in the training data, just no case has it.
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.