Skip to content

Commit 21d4975

Browse files
igerberigerber
authored
Merge pull request #451 from igerber/docs/goodman-bacon-2021-review
Add Goodman-Bacon (2021) paper review
2 parents e67bbc0 + 70795a6 commit 21d4975

2 files changed

Lines changed: 334 additions & 0 deletions

File tree

‎TODO.md‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -74,6 +74,7 @@ Deferred items from PR reviews that were not addressed before merge.
7474

7575
| Issue | Location | PR | Priority |
7676
|-------|----------|----|----------|
77+
| BaconDecomposition: substantive methodology audit pass against Goodman-Bacon (2021). Paper review on file at `docs/methodology/papers/goodman-bacon-2021-review.md` includes a proposed Methodology Registry Entry, but the `REGISTRY.md` `## BaconDecomposition` section (lines ~2598-2654) still carries the older contract. Audit pass needs to (a) verify `bacon.py` matches Theorem 1 / Eqs. 10a-g exactly, (b) decide how to handle genuinely always-treated units (`0 < first_treat <= min(time)`) — paper convention puts them in `U`, but `bacon.py` currently treats only `first_treat ∈ {0, np.inf}` as the `U` bucket, (c) generate R parity fixtures via `bacondecomp::bacon()`, (d) write `tests/test_methodology_bacon.py`, (e) replace the REGISTRY entry with the proposed text, (f) populate Verified Components / Corrections Made / Deviations in `METHODOLOGY_REVIEW.md` and flip status to Complete. | `diff_diff/bacon.py`, `docs/methodology/REGISTRY.md`, `tests/test_methodology_bacon.py`, `METHODOLOGY_REVIEW.md` | follow-up | Medium |
7778
| dCDH: Phase 1 per-period placebo DID_M^pl has NaN SE (no IF derivation for the per-period aggregation path). Multi-horizon placebos (L_max >= 1) have valid SE. | `chaisemartin_dhaultfoeuille.py` | #294 | Low |
7879
| dCDH: Survey cell-period allocator's post-period attribution is a library convention, not derived from the observation-level survey linearization. MC coverage is empirically close to nominal on the test DGP; a formal derivation (or a covariance-aware two-cell alternative) is deferred. Documented in REGISTRY.md survey IF expansion Note. | `chaisemartin_dhaultfoeuille.py`, `docs/methodology/REGISTRY.md` | #408 | Medium |
7980
| dCDH: Parity test SE/CI assertions only cover pure-direction scenarios; mixed-direction SE comparison is structurally apples-to-oranges (cell-count vs obs-count weighting). | `test_chaisemartin_dhaultfoeuille_parity.py` | #294 | Low |

0 commit comments

Comments
 (0)