Open
Conversation
…scope, saving CI)
S1: Fix double-dereference ``fuzzy'' / ``weights'' -> `fuzzy' / `weights'
S7: Read r(pilot_den) and r(h_den) into locals immediately after Mata call
before intervening commands can clear r() results
S8: Track Mata loading success via local mata_loaded instead of ambient _rc
S18: Remove redundant tempvar drops in r3d_bwselect.ado
S21: saving() now writes bootstrap CB bounds instead of pointwise normal CI
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- F1: extract wTplus/wTminus from T-regression Fortran output; use T-bandwidth weights (not outcome weights) when indexing e2_mat rows - F2: add positivity guards for user-supplied bandwidths in all three validation branches (fuzzy list, sharp list, direct scalar) - F3: remove duplicate out\$w_plus / out\$w_minus assignments (dead code) - F4: remove duplicate top-level out\$X, out\$Y_list, out\$T to avoid double memory use; update summary.r3d and print.r3d to read from object\$inputs\$X instead - F5: replace length-branching test validation with a single match.arg(..., several.ok=TRUE) call for uniform partial matching - F8: add IF (INFO .NE. 0) RETURN after each DGETRS call in locweights.f so mid-loop LAPACK failures surface correctly - Tests: add four regression tests covering F1/F2/F4/F5 findings Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Automated fixes from code review pipeline.
Summary
wTplus/wTminus) fore2_matrow indices instead of outcome weights — prevents wrong observation selection whenh_dendiffers fromh_num> 0) for all user-supplied bandwidth paths — prevents divide-by-zero in Fortranout$w_plus/out$w_minusassignments (dead code from copy-paste)out$X/out$Y_list/out$Ttop-level fields (already in$inputs); updatesummary.r3dandprint.r3dto read from$inputs$Xtestvalidation withmatch.arg(..., several.ok=TRUE)for uniform partial matching at any vector lengthIF (INFO .NE. 0) RETURNafter eachDGETRScall inlocweights.fso mid-loop LAPACK failures surface rather than being silently overwrittenTest plan
devtools::check()passes with 0 errors, 0 warnings (4 pre-existing NOTEs)h_den < h_numverifiese2_matrows outside T-bandwidth are zerobandwidths=list(num=0)/bandwidths=-0.5/den=0all produce clear error$X/$Y_list/$Tabsent; accessible via$inputstest=c("null","homo")partial matching works same as full names🤖 Generated with Claude Code