BUG: Change (n+1) to n for correct jackknife calculation of hd quantile standard error #16481
Conversation
This file contains 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
…ile standard error
|
Thanks @SamGRosen. The correction looks right to me. I've suggested a refactoring of your test in SamGRosen#1. If you agree with and merge that, I think this is good to go. |
TST: mstsats.hdquantiles_sd: refactor test per Efron
|
@mdhaber LGTM! Done. |
|
Thanks @SamGRosen! |
Giodiro
pushed a commit
to Giodiro/scipy
that referenced
this pull request
Jun 29, 2022
…le standard error (scipy#16481) * BUG: FChange (n+1) to n for correct jackknife calculation of hd quantile standard error Co-authored-by: Matt Haberland <mhaberla@calpoly.edu>
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Outdated suggestions cannot be applied.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.
Reference issue
N/A
What does this implement/fix?
In Harrell and Davis' original paper (Formula 4), the standard error calculation for the quantiles is a jackknife estimator. A jackknife estimator requires calculating the estimator in a leave-one-out fashion; however, the current implementation uses
(n+1)when calculating the weights used in the jackknife, when it should usen, as the jackknife leaves 1 data point out.Appropriate unit tests updated. Changes include testing against the new correct values, and doing a rudimentary jackknife calculation to compare to the actual method.
Additional information
Funny enough, Harrell actually makes the same mistake in his implementation, which I'm guessing was used to verify correctness of the previous one.