Conversation
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Closes #243.
step_lencode_glm()set the value used for novel/unseen levels to a trimmed mean of the per-level coefficients, which gives every level equal weight regardless of how many observations it has. For a factor with one dominant level and many small ones, this pushed the novel-level value far away from the outcome as a whole.The value is now the coefficient of an intercept-only GLM fit to the whole outcome (weighted the same way as the per-level model), so it reflects the overall/global outcome rather than an unweighted average across levels.
step_lencode_bayes()andstep_lencode_mixed()have the same issue (they also fall back to an unweighted average/trim of per-level effects for..new) but are out of scope here.