Repository navigation
Use dimension-adjusted GVIFs for collinearity plots and severity groups - #998
Open
jbogomolovas2 wants to merge 1 commit into
Open
jbogomolovas2 wants to merge 1 commit into
jbogomolovas2 wants to merge 1 commit into
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Hi, I noticed that
performance0.19.0.3 correctly accounts for term degrees of freedom in theadj. VIFcolumn, but the plots and severity groups still compare raw GVIFs with the usual 5/10 cutoffs.I suggest using
GVIF^(1/df)for these displays, keeping the existing logarithmic axis and 5/10 bands. For example, a term with 11 coefficients and raw GVIF of 2048 would be displayed at2048^(1/11) = 2.This is simply the square of the existing adjusted GVIF:
For one-coefficient terms, this recovers ordinary VIF exactly. The same normalization is used by
semEff::VIF(). Keeping 5/10 therefore preserves the existing convention on a variance-inflation scale.This PR updates severity groups, the interaction warning threshold, and plotting data, including consistently transformed confidence limits. Raw GVIFs, confidence limits, and adjusted GVIFs in the returned table remain unchanged.
The companion
seePR labels the plotted scale explicitly.Validation: 313 targeted expectations passed; 21 optional/CRAN-skipped tests. Regression coverage includes factor interactions, confidence limits,
check_model()consistency, and unchanged one-coefficient results.