Skip to content

fixes & enhancements relating to null covar of imbalances, svg's from base R plot methods#143

Open
benthestatistician wants to merge 12 commits intomarkmfredrickson:mainfrom
benbhansen-stats:master
Open

fixes & enhancements relating to null covar of imbalances, svg's from base R plot methods#143
benthestatistician wants to merge 12 commits intomarkmfredrickson:mainfrom
benbhansen-stats:master

Conversation

@benthestatistician
Copy link
Collaborator

Adjustments having to do with idiosyncratic use cases. No rush about review or pull.

also avoids error in cases with an NA in the std.diff column
... I think.  To be be tested by CI but I can't confirm locally
namely, edge case that variable names of bal plot is a length-0 vector
a crude adjustment only
As in reported balanceTest results, intercepts and not-missing indicators
that only ever take the value 1 get removed. Now this is also done with
the null covariance of balance comparisons that is returned invisibly as
an attribute.
... are now documented in balT inline comments, one test
W/ a bit more to be done to make corresp p-vals equal NA not 1.
Adding but commenting out tests for that behavior.
With const variables, overall p-value is now also NA
Deliver message not warning
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant

Comments