Fix FAST CI estimation via bootstrapping from contiguous blocks#652
Open
ConnectedSystems wants to merge 2 commits intomainfrom
Open
Fix FAST CI estimation via bootstrapping from contiguous blocks#652ConnectedSystems wants to merge 2 commits intomainfrom
ConnectedSystems wants to merge 2 commits intomainfrom
Conversation
ConnectedSystems
changed the title
Member
Author
|
Tests are failing due to an outdated workflow config. #653 has the necessary changes to resolve the issue so once that's merged in I'll rebase this PR. |
The method better preserves local frequency structure - sampling with replacement was too naive an approach.
ConnectedSystems
force-pushed
the
correct-FAST-bootstrap
branch
from
df061f7 to
badeff0
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Correct the bootstrapping method applied for FAST analysis by replacing the sampling with replacement approach with a block boostrapping approach which, in my view, better preserves the local frequency structure.
Closes #649 (thanks to @SchnFlo for suggesting the fix).
In #649, two potential methods are proposed to resolve the bug.
The first (method 1) is to first sort the samples, producing results that were reportedly more in line with the Sobol' confidence bounds.
The second (method 2) was to apply the block boostrapping approach.
Ultimately, I decided to go with the block bootstrap (method 2) as I think it aligns more with the FAST analysis method (which is based on frequency analysis).