Commits

History for

Commits on Apr 19, 2026

Address CI AI review: NaN first_treat + StaggeredTripleDifference coverage

igerber
and
claude
committed
65d1bd8
Address CI AI review: reject negative first_treat in ContinuousDiD

igerber
and
claude
committed
a036525
Address CI AI review: close remaining axis-E sites + warning accuracy

igerber
and
claude
committed
e5755c2
Surface row-count for axis-E silent coercion / drop sites

igerber
and
claude
committed
a5f38e1

Commits on Mar 28, 2026

Preserve df=0 sentinel on DoseResponseCurve for NaN serialization

igerber
and
claude
committed
b26eeaa
Complete replicate guard coverage and fix remaining review items

igerber
and
claude
committed
8b8dcb3
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
ab5377f
Fix replicate df override: only when replicates actually dropped

igerber
and
claude
committed
06cf61b
Add positive-weight effective sample gates and fix replicate df metadata

igerber
and
claude
committed
81ef2f5

Commits on Mar 27, 2026

Fix CS df key name, re-read df after aggregation, propagate ContinuousDiD df

igerber
and
claude
committed
5930286
Return n_valid from replicate variance functions, fix df properly

igerber
and
claude
committed
84809cd
Guard <2 valid replicates, drop logit columns, filter NaN cells

igerber
and
claude
committed
50d8314
Guard zero-weight cells in TripleDiff and ContinuousDiD

igerber
and
claude
committed
657d918
Fix replicate IF variance score scaling for EfficientDiD, TripleDiff, ContinuousDiD

igerber
and
claude
committed
b063f5e
Fix TripleDiff replicate IF scale, reject replicate+bootstrap, harden DEFF

igerber
and
claude
committed
2cd91b9
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
Remove allow_zero_se — let SE=0 produce NaN inference consistently

igerber
and
claude
committed
1b5c2f0
Wire allow_zero_se through survey bootstrap callers, add TROP deviation note

igerber
and
claude
committed
f23548f

Commits on Mar 25, 2026

Address CI review P0/P1 findings for PR #237

igerber
and
claude
committed
1aa6d16

Commits on Mar 24, 2026

Address round-3 AI review findings for survey bootstrap

igerber
and
claude
committed
c0f6e37
Address AI review P0/P1/P2 findings for survey bootstrap

igerber
and
claude
committed
e254bef
Add survey-aware bootstrap for all estimators (Phase 6)

igerber
and
claude
committed
7b2c9c8

Commits on Mar 22, 2026

Fix DoseResponseCurve survey df and strengthen P3 tests from PR #226 review (round 10)

igerber
and
claude
committed
f3d7427
Fix ContinuousDiD fweight TSL scaling and BaconDecomposition empty-cell guard from PR #226 review (round 6)

igerber
and
claude
committed
e8fcf88
Fix ContinuousDiD bread normalization and panel-level survey metadata from PR #226 review (round 4)

igerber
and
claude
committed
260728a
Fix weighted-mass IF linearization in ContinuousDiD and TripleDifference from PR #226 review (round 3)

igerber
and
claude
committed
ce49f70
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 Feb 22, 2026

Fix not-yet-treated control mask to respect anticipation parameter

igerber
and
claude
committed
999da34
Fix bootstrap NaN propagation for rank-deficient cells and remove unused param

igerber
and
claude
committed
a8c1c2e
Fix NaN propagation in rank-deficient spline predictions

igerber
and
claude
committed
55b0d2d