FazBrowse GitHub Viewer | Trending | | Home

Tools: [Download Repo ZIP] [Original HTTPS Page]

History for diff_diff/efficient_did.py - UtilityTools/diff-diff · GitHub

Commits

History for

Commits on Apr 19, 2026

Address eleventh round of CI review findings on PR #318

igerber
and
claude
committed
4bb7dcd

Commits on Apr 4, 2026

Extend zero-weight never-treated guard to nocov path, fix roadmap inconsistency

igerber
and
claude
committed
ca9522c
Restore replicate-weight dispatch in aggregation, raise on zero-weight never-treated

igerber
and
claude
committed
42d3dff
Fix IF-scale mismatch in survey aggregation, bootstrap scores, zero-weight guards

igerber
and
claude
committed
aca9f8b
Address P1/P2 review findings: survey-weighted WIF, bootstrap denominator, zero-weight guard

igerber
and
claude
committed
0a12a77
Add survey design support to EfficientDiD covariates (DR) path

igerber
and
claude
committed
2205914

Commits on Mar 28, 2026

Validate replicate scale/rscales, combined_weights contract, fix df metadata

igerber
and
claude
committed
6c6494f
Guard all replicate-df paths for NaN inference when rank <= 1

igerber
and
claude
committed
dd8f51b
Guard all replicate-df paths for NaN inference when rank <= 1

igerber
and
claude
committed
ab5377f
Add positive-weight effective sample gates and fix replicate df metadata

igerber
and
claude
committed
81ef2f5

Commits on Mar 27, 2026

Propagate replicate df through EfficientDiD and TripleDifference IF paths

igerber
and
claude
committed
86f09f2
Return n_valid from replicate variance functions, fix df properly

igerber
and
claude
committed
84809cd
Fix subpopulation mask validation, EfficientDiD bootstrap guard, logit rank action

igerber
and
claude
committed
2c2ef14
Fix replicate IF variance score scaling for EfficientDiD, TripleDiff, ContinuousDiD

igerber
and
claude
committed
b063f5e
Fix panel→unit replicate propagation, weight normalization, validation

igerber
and
claude
committed
0a11b76

Commits on Mar 26, 2026

Address AI review P1/P2 findings for replicate weights

igerber
and
claude
committed
7bd49f2
Add survey Phase 6: replicate weights, DEFF diagnostics, subpopulation analysis

igerber
and
claude
committed
29eae1e

Commits on Mar 24, 2026

Add survey-aware bootstrap for all estimators (Phase 6)

igerber
and
claude
committed
7b2c9c8

Commits on Mar 22, 2026

Add WIF correction to Hausman ES(e) EIF, fix doc inconsistencies

igerber
and
claude
committed
56beb42
Rewrite Hausman pretest to ES(e) per Theorem A.1, guard cluster+survey

igerber
and
claude
committed
bf427af
Filter NaN effects in Hausman pretest, add clustered aggregation tests

igerber
and
claude
committed
94be0c0
Add empty-treatment guard, last_cohort aggregate/bootstrap tests, fix n_obs

igerber
and
claude
committed
dd72e0f
Fix clustered SE formula, last_cohort semantics, Hausman n_units, add cluster validation

igerber
and
claude
committed
11f7e23
Address P2 review findings: remove mutable state, deduplicate cluster covariance, update docstrings

igerber
and
claude
committed
9e02af6
Fix stale cluster indices in Hausman pretest row filtering

igerber
and
claude
committed
fdc86fe
Add cluster-robust SEs, last-cohort control, Hausman pretest, small cohort warning

igerber
and
claude
committed
028a890
Reject EfficientDiD covariates + survey_design combination from PR #226 review (round 11)

igerber
and
claude
committed
4f04832
Fix EfficientDiD fweight consistency and BaconDecomposition index safety from PR #226 review (round 5)

igerber
and
claude
committed
719bfec
Fix ContinuousDiD bread normalization and panel-level survey metadata from PR #226 review (round 4)

igerber
and
claude
committed
260728a
Fix 4 P1 issues from PR #226 review (round 2)

igerber
and
claude
committed
a8a6d2c
Fix 5 P1 issues from PR #226 review (round 1)

igerber
and
claude
committed
f823a7a
Add Phase 3 survey design support for 6 OLS-based estimators

igerber
and
claude
committed
42590aa

Commits on Mar 21, 2026

Add overlap warning for inverse propensity clipping, align module docstring

igerber
and
claude
committed
a3a7909
Document OLS outcome regression as working-model choice in REGISTRY/docstrings

igerber
and
claude
committed
56fa2e7
Implement sieve inverse propensity for conditional Omega* (Eq 3.12 step 4)

igerber
and
claude
committed
16dc2f9

Back | FazBrowse Home | New Git URL