v4.0.0

Breaking Changes

• LWDiDResults.pvalue and LWDiDResults.ci are removed (ledger rows M-151, M-152): the
  3.10-born alias properties for p_value / conf_int were first shimmed with a FutureWarning
  and then removed outright at 4.0 (item 9b of the cut). Read .p_value / .conf_int.

• LWDiDResults.to_dataframe() is reshaped onto the shared results schema (ledger row
  M-153): each level returns aggregate(level).to_dataframe() - columns level, label, target, att, se, t_stat, p_value, conf_int_lower, conf_int_upper, n, weight, df (the
  event-study schema for level="event_study") - plus LWDiD's rolling, estimation_method
  and vcov_type columns. The default "simple" is ONE headline row; the former per-cohort
  rows are level="group" (without the old trailing "Overall" row). The term / cohort
  column is now label, n_treated is the schema's n, n_obs and n_control are no longer
  in the frame (both stay in to_dict()), and ci_lower / ci_upper are
  conf_int_lower / conf_int_upper. to_csv() follows and takes the same level.

• BusinessReport.to_dict() schema 2.0 -> 3.0: the StackedDiD assumption block no longer
  carries the legacy clean_control key (read control_group). DiagnosticReport's schema stays
  2.0 at this step. (Intermediate: later bullets in this release move the final 4.0 versions to
  BUSINESS_REPORT_SCHEMA_VERSION 4.0 and DIAGNOSTIC_REPORT_SCHEMA_VERSION 3.0.)

• Passing aggregate= or balance_e= to CallawaySantAnna.fit, EfficientDiD.fit,
  StackedDiD.fit, ChaisemartinDHaultfoeuille.fit or HeterogeneousAdoptionDiD.fit (or
  aggregate= to did_had_pretest_workflow) now raises Python's native
  TypeError: ... unexpected keyword argument. TripleDifference's staggered-mode
  aggregate=/balance_e= are untouched at this step (M-140/M-141; they are REMOVED later in
  this release by the DDD container unification bullet below).

• Positional slots after the removed parameters shift left. CallawaySantAnna:
  survey_design 9 -> 7; EfficientDiD: survey_design 9 -> 7, store_eif 10 -> 8;
  StackedDiD (aggregate sat before them): population, survey_design, covariates
  7/8/9 -> 6/7/8; ChaisemartinDHaultfoeuille: L_max, covariates, trends_linear,
  trends_nonparam, honest_did, heterogeneity, design2, survey_design each move one
  slot left (7..14 -> 6..13). HeterogeneousAdoptionDiD.fit shifts nothing (everything after
  the removed slot is keyword-only; a seventh positional now raises). Pass these by keyword.

• CallawaySantAnnaResults.to_dataframe(level="event_study"|"group") and the
  EfficientDiDResults twins raise on every fresh fit (the legacy fields are always None);
  use results.aggregate(...).to_dataframe(). (Intermediate: the M-093 schema bullet below
  supersedes this - since item 8b to_dataframe(level="event_study") recomputes the surface on
  these classes instead of raising; level="group" still raises with the post-fit pointer.)

• Passing aggregate= or balance_e= to ImputationDiD.fit, TwoStageDiD.fit or
  ContinuousDiD.fit now raises Python's native TypeError: ... unexpected keyword argument.

• Positional slots after the removed parameters shift left. ImputationDiD and
  TwoStageDiD: survey_design 9 -> 7; ContinuousDiD: survey_design, covariates
  8/9 -> 7/8. Pass these by keyword.

• aggregate('event_study', balance_e=k) on a BOOTSTRAPPED ImputationDiD / TwoStageDiD
  fit raises NotImplementedError (the relayed surface was computed unrestricted; re-fit with
  n_bootstrap=0 for a balanced window). Analytic fits recompute with balance_e as before.

• A bootstrapped ImputationDiD / TwoStageDiD / ContinuousDiD fit WITH bootstrap inference
  but WITHOUT a stored surface - a 3.9-3.13 bootstrapped plain-fit pickle, or a fit whose
  fail-soft materialization dropped it - raises ValueError from aggregate('event_study') /
  aggregate('group') (previously NotImplementedError from the bootstrap gate); the message
  names the refit. A 3.9-3.13 bootstrapped pickle fitted with balance_e=k stores a restricted
  surface and the relay returns it as stored.

• TripleDifference.fit(time=) in the 2x2x2 design now raises ValueError naming post=
  (M-085). Through 3.x time= was the deprecated alias for the 0/1 post indicator and emitted
  a FutureWarning (M-031); since 4.0 time means the staggered calendar column only.
  Supplying both post= and time= keeps raising the "pass only post=" error.

• Keyword mapping for callers of the removed class: eligibility= is partition= on
  TripleDifference; the staggered fit parameters are keyword-only; cluster= raises in
  staggered mode (use n_bootstrap > 0 for unit-level clustered inference). Users of
  balance[did] stay on 3.x through that extra's diff-diff<4 pin, where the old name still
  resolves.

• TwoWayFixedEffects.fit(time=) in static mode now raises ValueError naming post=
  (M-083). Through 3.x time= was the deprecated alias for the 0/1 post indicator and
  emitted a FutureWarning (M-082); since 4.0 time means the event-study calendar column
  only. Supplying both post= and time= keeps raising the "pass only post=" error.

• EventStudyResults.period_effects is removed (M-016; the compatibility view
  this item first shipped was retired at 4.0 by item 9b - no release carried it). Through
  the pre-release window, on a calendar surface it returned the former calendar-keyed dict of
  per-period records (period, effect, se, t_stat, p_value, conf_int, is_significant, significance_stars, reference period omitted; is_significant is evaluated at the fit's
  alpha, where the removed record hard-coded 0.05; duplicate labels raise). Relative-time
  aggregate("event_study") surfaces do not carry it (AttributeError, no warning). The
  successor is to_dataframe() and the columnar arrays.

• The 3.x "default reference_period has changed" FutureWarning is no longer emitted
  (M-007): the event-study mode defaults to the last pre-period silently.

• simulate_power and friends read the flat att/se/p_value/conf_int names from
  unregistered custom estimators' results and raise TypeError naming _EstimatorProfile
  otherwise (the avg_* fallback left with the removed container; pass result_extractor=
  for other shapes).

• 4.0 cut, storage flip of the headline quintet (M-050..M-058). CallawaySantAnnaResults (and DMLDiDResults), SunAbrahamResults, ImputationDiDResults, TwoStageDiDResults, StackedDiDResults, EfficientDiDResults, WooldridgeDiDResults, ChaisemartinDHaultfoeuilleResults and ContinuousDiDResults now store the canonical att / se / t_stat / p_value / conf_int as their native dataclass fields (renamed in slot, so the positional constructor layout is unchanged) and take those names as constructor keywords (overall_att= raises TypeError). The 3.x overall_att / overall_se / overall_t_stat / overall_p_value / overall_conf_int names - and on ContinuousDiDResults all nine legacy spellings, the overall_att_* fields and the former silent overall_se-style aliases - are removed at 4.0 (item 9b retired the FutureWarning property shims this item first shipped - reading one raises AttributeError); they are absent from dataclasses.fields(), asdict() and to_dict(). 3.x pickles load unchanged (each container's __setstate__ rewrites the legacy keys on load). The bootstrap draw containers (bootstrap_results.overall_att_se and siblings) are unchanged; StaggeredTripleDiffResults kept its overall_* storage at this step and is removed by the DDD container unification bullet below.

• Report schemas. DiagnosticReport.to_dict() reports headline_metric.name and target_parameter.headline_attribute as "att" for the nine containers above ("overall_att" under 3.x); DIAGNOSTIC_REPORT_SCHEMA_VERSION 2.0 -> 3.0. BusinessReport.to_dict() reports target_parameter.headline_attribute likewise; BUSINESS_REPORT_SCHEMA_VERSION 3.0 -> 4.0 (its headline block carries no name key). StaggeredTripleDiffResults kept headline_attribute: "overall_att" at this step (removed by the DDD unification below).

• 4.0 cut, DDD container unification (M-014, M-140, M-141). TripleDifference.fit() returns ONE TripleDifferenceResults for both designs: results.design is "2x2x2" or "staggered", the canonical att / se / t_stat / p_value / conf_int are native in both, the 2x2x2 positional layout is unchanged, and the staggered structure (group_time_effects, groups, time_periods, unit counts, control_group, base_period, anticipation, bootstrap_results, ...) is appended with defaults (group_time_effects is {} in the 2x2x2 design). The 3.x StaggeredTripleDiffResults class, its module, its overall_* fields and to_dict() keys, and its fit-time overall_att_es / overall_se_es / overall_t_stat_es / overall_p_value_es / overall_conf_int_es fields are REMOVED (3.x staggered pickles cannot load; 3.x 2x2x2 pickles load unchanged). The fit-time aggregate= / balance_e= keywords of TripleDifference.fit() are REMOVED (TypeError): aggregation is post-fit - results.aggregate('event_study') (EventStudyResults, balance_e=), results.aggregate('group') (AggregationResult) and results.aggregate('simple') (a one-row relay of the stored headline, both designs); bootstrapped fits REPLAY the fit-time multiplier bootstrap so sup-t bands and percentile inference match a 3.x fit-time aggregation; the 2x2x2 design fails the staggered levels closed. The paper's Eq. 4.14 overall is results.aggregate('simple', weights='event_time') (accepts balance_e=; both 3.x warnings preserved). summary() no longer prints the event-study / cohort / Eq. 4.14 blocks; to_dataframe('event_study' | 'group') raise with the post-fit pointer; to_dataframe(level=None) defaults to 'group_time' in the staggered design and 'overall' in 2x2x2 (whose one-row frame gains design, pscore_trim, alpha, df_inference). New field df_inference records the 2x2x2 engine's t df. plot_group_effects and its siblings raise ValueError on a 2x2x2 fit (no cells).

• Report schemas. For staggered DDD fits DiagnosticReport.to_dict()['estimator'] and BusinessReport.to_dict()['estimator']['class_name'] read TripleDifferenceResults, and target_parameter.headline_attribute reads "att" (the last "overall_att"). Both schemas gain an additive key - DR a top-level design, BR estimator.design - "2x2x2" / "staggered" for TripleDifferenceResults, None otherwise. No version bump: DIAGNOSTIC_REPORT_SCHEMA_VERSION 3.0 and BUSINESS_REPORT_SCHEMA_VERSION 4.0 had not shipped (folded into the PR 5b bump). power.py's staggered-boundary guard no longer keys on aggregate / balance_e (an estimator_kwargs carrying them surfaces as fit()'s TypeError).

• 4.0 cut, df_convention default flip (M-004, M-005, M-006, M-128, M-129, M-130, M-131). The analytical tail-df convention for CLUSTERED fits defaults to "cluster" (the Stata/fixest t(G - 1) reference) instead of "residual" (the fitted residual df t(n - K_full)) on DifferenceInDifferences, TwoWayFixedEffects, LinearRegression, SunAbraham, WooldridgeDiD (OLS arms), StackedDiD and ImputationDiD (pretrends leads); LPDiD already defaulted to "cluster" at 3.9. What moves: p-values and confidence intervals on clustered analytical fits - TwoWayFixedEffects (static, and event-study with unit=), SunAbraham, WooldridgeDiD OLS and StackedDiD cluster at unit by default, so their default fits move; ImputationDiD moves on the pretrends=True lead rows; DifferenceInDifferences / LinearRegression move only with an explicit cluster= / cluster_ids=; the analytical lanes inside bootstrapped fits (WooldridgeDiD per-cell inference and post-fit aggregate(), ImputationDiD bootstrapped leads) move too, as do the fake-timing / fake-group placebo diagnostics and the power simulators that relay those p-values. What does not: point estimates, standard errors and t-statistics; survey-design fits (survey df keeps precedence); vcov_type="hc2_bm" (Bell-McCaffrey DOF keep precedence); bootstrap percentile p-values / CIs; unclustered and Conley fits; the BJS overall/post inference of ImputationDiD; WooldridgeDiD logit/poisson arms; permutation and leave-one-out diagnostics. df_convention="residual" restores the 3.x inference bit-for-bit; on the surfaces where the knob is a documented no-op (non-pretrends ImputationDiD, WooldridgeDiD logit/poisson) the inert-config UserWarning now fires for an explicit "residual" rather than for an explicit "cluster". SyntheticDiD inherits the flipped attribute from DifferenceInDifferences without reading it (no behaviour change).

• SpilloverDiD auto-clusters at unit when cluster= is omitted ([M-080], item 7b of the 4.0 cut), mirroring TwoWayFixedEffects' carve-outs (explicit cluster= wins; the auto-cluster is dropped on the Conley path and never injected as a survey PSU). Its default standard errors, p-values and CIs move; cluster=False reproduces the 3.x observation-level HC1 numbers. Results report cluster_name="unit" / n_clusters. Inference stayed on the residual df at this step; the df_convention knob (next bullet, [M-154]) moves it to t(G - 1) in the same release.

• cluster=False is the uniform explicit off-switch ([M-080]): the plain heteroskedasticity-robust sandwich with residual-df t inference on TwoWayFixedEffects (static + event-study; one-way CR2 Bell-McCaffrey under hc2_bm), SunAbraham, WooldridgeDiD (OLS within-transform lane with the absorbed-FE full-K rescale; logit/poisson plain QMLE sandwich) and LPDiD (per-horizon HC1; if_robust unclustered IF variance on the regression-adjustment path; implicit per-observation PSUs under a survey design); an accepted bit-identical no-op on the unit-level-by-construction estimators (CallawaySantAnna incl. its RC auto-PSU path, EfficientDiD incl. hausman_pretest, DMLDiD, ChaisemartinDHaultfoeuille, staggered TripleDifference, LWDiD incl. the sensitivity helpers, HeterogeneousAdoptionDiD, ImputationDiD, TwoStageDiD) and on the 2x2 DifferenceInDifferences / 2x2x2 TripleDifference surfaces; still rejected on StackedDiD. Rejected with inference="wild_bootstrap" (DiD/TWFE), WooldridgeDiD(n_bootstrap > 0) and vcov_type="hc2_bm" on SunAbraham/WooldridgeDiD (cluster-only Bell-McCaffrey DOF; tracked in TODO.md).

• cluster accepts str | None | False only ([M-155], diff_diff.utils.validate_cluster): True and non-str column labels raise one shared ValueError at construction, through set_params and at fit() on every estimator exposing the parameter (DMLDiD's existing rule generalized; integer-labelled cluster columns must be renamed).

• Cluster metadata: DifferenceInDifferences / TwoWayFixedEffects report n_clusters from the regression's effective cluster count on non-survey analytical fits (previously None unless the wild bootstrap ran); SunAbrahamResults gains an appended n_clusters field and reports cluster_name on clustered analytical fits (previously None off the Conley path), and its summary() renders the variance-family label on analytical non-survey fits; TwoWayFixedEffects under a survey design with no explicit cluster= reports cluster_name=None (the auto-cluster is never injected as a PSU).

• SpilloverDiD gains the three-value df_convention knob with the 4.0 default "cluster" ([M-154], item 7c of the 4.0 cut): p-values and confidence intervals on its clustered lanes - the unit auto-cluster and an explicit cluster= - now use the Stata/fixest t(G - 1) reference like every other auto-clustered default, with G the effective cluster count reported as n_clusters. Point estimates, standard errors and t-statistics never move. df_convention="residual" restores the 3.x residual-df inference bit-for-bit; "normal" is deliberate z. "cluster" falls back to the residual df on cluster=False and on every vcov_type="conley" fit, an explicit product-kernel cluster= included ("normal" still selects z there); a survey_design df keeps precedence under every value; a saturated stage-2 sample stays all-NaN under every value. SpilloverDiDResults reports df_convention and inference_df (also in to_dict()).

• Event-study reference rows are marked explicitly ([M-093], item 8a of the 4.0 cut): every native event_study_effects dict that materializes its normalization row (StackedDiD, ImputationDiD, TwoStageDiD, SpilloverDiD, the CallawaySantAnna/DMLDiD universal-base aggregation carrier) now carries "is_reference": True on that row and NO count key there - the 3.x n_obs == 0 / n_groups == 0 sentinel is retired, and every remaining zero count is a genuine count (a non-estimable horizon). SpilloverDiD's event-study att_dynamic / spillover_effects frames gain a boolean is_reference column with n_obs = NaN on the reference rows (the column is float; to_dict() records follow). The native to_dataframe(level="event_study") frames of StackedDiD / bootstrapped ImputationDiD / TwoStageDiD rendered the reference row's n_obs as NaN (float column) at this step; the 8b schema rewire (next bullet) replaces those frames. plot_event_study, plot_honest_event_study, HonestDiD, PreTrendsPower and DiagnosticReport locate the reference by the key on their raw-dict routes - dispatch, admission, rendering and exception types are unchanged. TwoStageDiD.summary() now prints its genuine all-filtered horizons as NaN rows instead of a mislabeled [ref: h] line. 3.x pickles are migrated on load. SunAbraham (omitted row, reference_period) and EfficientDiD (PT-Post anchor, an estimated cell named by reference_period) are unchanged.

• Every results class's to_dataframe(level="event_study") now emits the unified EVENT_STUDY_SCHEMA ([M-093], item 8b of the 4.0 cut): event_time, att, se, t_stat, p_value, conf_int_lower, conf_int_upper, cband_lower, cband_upper, n, df, is_reference, estimand - the frame results.aggregate("event_study").to_dataframe() returns (CallawaySantAnna/DMLDiD, EfficientDiD, ImputationDiD, TwoStageDiD, StackedDiD, ContinuousDiD, TripleDifference staggered, ChaisemartinDHaultfoeuille, HAD event-study results, DurationDiD, LWDiD) or the builder frame (SunAbraham, LPDiD, WooldridgeDiD, SpilloverDiD). The 3.x bespoke frames are gone: relative_period -> event_time; effect / att_glob / LPDiD's coefficient -> att; conf_int_lo/hi, conf_low/high, HAD's conf_int_low/high -> conf_int_lower/conf_int_upper; n_obs -> n (NaN on the is_reference row); dCDH's horizon -> event_time on the paper's l1_first_switch convention, its per-horizon estimand labels (DID^pl_l / ref / DID_l) are not carried (estimand is "att") and its reference row is marked is_reference instead of n_obs 0. WooldridgeDiD and LWDiD append their fit-configuration columns after the schema (cohort_trends, aggregation_weights; rolling, estimation_method, vcov_type), and WooldridgeDiD.to_dataframe(level="event_study") (its default level) now RAISES before aggregate("event_study", weights=) instead of returning an empty frame. CallawaySantAnna, EfficientDiD, TripleDifference and analytical ImputationDiD / TwoStageDiD / ContinuousDiD no longer raise "not stored on 4.0 fits" - the level recomputes the surface from the retained kit (bootstrapped CallawaySantAnna / EfficientDiD / TripleDifference fits replay the fit-time bootstrap on every call; an ImputationDiD replicate-weight survey fit with pretrends=True raises NotImplementedError as aggregate() does); a kit-less pre-3.9 pickle, or a bootstrapped pickle predating the replay state, that still carries a stored dict now raises where it returned a frame. level="group" on those three classes still raises with the post-fit pointer. New levels: SpilloverDiDResults.to_dataframe(level="simple" | "event_study") (default "simple", the former one-row frame; aggregate-mode fits raise on "event_study"), HeterogeneousAdoptionDiDEventStudyResults.to_dataframe(level="event_study"), DurationDiDResults.to_dataframe(level="event_study"). An LPDiD fit whose stored event-study frame is empty raises instead of returning the empty frame.

• 4.0 cut, item 9b: the compatibility shims scheduled for a 5.0 removal are removed at 4.0 (maintainer
  decision - no 4.x debt; none of these FutureWarning shims ever reached a release, so the released
  3.x names break without a warned release - docs/migration-4.0.md is the notice). The 3.x
  overall_att / overall_se / overall_t_stat / overall_p_value / overall_conf_int names on
  CallawaySantAnnaResults (and DMLDiDResults), SunAbrahamResults, ImputationDiDResults,
  TwoStageDiDResults, StackedDiDResults, EfficientDiDResults, WooldridgeDiDResults,
  ChaisemartinDHaultfoeuilleResults and - all nine spellings, overall_att_* included -
  ContinuousDiDResults raise AttributeError; read the canonical att / se / t_stat /
  p_value / conf_int (M-050..M-058). 3.x pickles still load (the __setstate__ key migration
  stays). EventStudyResults.period_effects is gone; read to_dataframe() or the columnar arrays
  (M-016). LWDiDResults.pvalue / .ci are gone; read .p_value / .conf_int (M-151, M-152).

• SyntheticControl(inner_max_iter=, inner_min_decrease=) are removed (M-148, M-149; the
  DeprecationWarning shipped in 3.13.0): passing either is a TypeError, and the positional
  parameters after them (standardize, alpha, seed, v_cv_t0) move two slots left - pass
  them by keyword.

• run_placebo_test / run_all_placebo_tests: time is the calendar column only and
  post_periods is required for every test_type (M-156, M-157; spec section 8 rule 1). The
  0/1 post dummy the permutation and leave-one-out tests need is derived as
  data[time].isin(post_periods); a missing time column, missing/empty post_periods, missing
  values in time, post_periods values absent from the column, or a pre-existing _placebo_post
  column raise ValueError (on the suite wrapper before any test runs). Calls that passed the 0/1
  indicator as time= with post_periods=[1] return identical numbers; calls that omitted
  post_periods must add it.

Added

• Python 3.15 support: requires-python now allows 3.15 (<3.16) and
  the 3.15 classifier is declared; release wheels are built for CPython 3.15
  on Linux (x86_64, aarch64), macOS arm64 and Windows, and CI runs the
  Rust-backend matrix and the pure-Python fallback on 3.15. The Rust
  extension builds against PyO3 0.30, which supports 3.15.0 final.
• LWDiDResults.summary(alpha=None) / print_summary(alpha=None) join the family-wide
  summary-alpha guard (ledger row M-146): a value other than the fit-time alpha raises
  ValueError instead of being ignored; the methods previously took no alpha at all.
• LWDiDResults.to_dataframe(level=...) / to_csv(path, level=...) gain the uniform
  level selector ("simple" default, "group", "event_study") - see Breaking Changes.
• DurationDiDResults.aggregate("simple"): a verbatim relay of the stored headline
  quintet (n = n_units, df NaN - inference is bootstrap-only), joining the spec section 6
  simple contract every aggregation adopter honours.
• RDDensityTestResult.to_dict(): JSON-serializable scalars with native casts (NaN
  inference passes through), the *_conventional block None unless report_all=True, and
  the bandwidth-selection table nested under bandwidths only when it was estimated.

Changed

• WooldridgeDiDResults.summary(alpha=None) renders the headline (simple) row only; the
  group / calendar / event-study tables it used to render on request - and their [pre]/[antic]
  relative-period labels - are no longer part of summary(). Use aggregate(...) /
  to_dataframe(level=...); anticipation is still reported on the results object.
  to_dataframe(level=) now rejects an unknown level with ValueError instead of returning an
  empty frame.
• practitioner_next_steps no longer recognises duck-typed results that carry only the legacy
  estimator attribute; ChangesInChangesResults.method is the routing field.
• Runtime messages, the four llms*.txt guides, the API docs, the migration guide and the
  benchmarks no longer steer users to the fit-time route on these five estimators;
  CallawaySantAnna's class docstring, plot_event_study, honest_did / pretrends and the
  results-class "not stored" messages point at results.aggregate('event_study').
• Bootstrapped ImputationDiD / TwoStageDiD / ContinuousDiD fits relay their event-study
  (and, on Imputation/TwoStage, group) surface. The deprecated keyword was the only route to
  bootstrap event-study inference there (no per-draw state is retained for a replay), so a fit
  with n_bootstrap > 0 now ALWAYS computes those surfaces at fit - exactly what
  fit(aggregate="all") computed, balance_e=None - stores them in the legacy
  event_study_effects / group_effects fields, and aggregate('event_study') /
  aggregate('group') RELAYS them (kit-snapshot alpha / anticipation / survey metadata; NaN
  df column; vcov None) instead of raising NotImplementedError. Analytic fits are
  unchanged (legacy fields None, kit recompute on demand). Storage rule: the legacy fields
  hold a surface IFF the fit carries bootstrap inference. On Imputation/TwoStage a bootstrapped
  fit that ends analytical (zero valid draws, total bootstrap failure) clears them and
  aggregate() recomputes analytically. ContinuousDiD's bootstrapped kit is scalars-only, so
  no analytical fallback exists there: a bootstrapped fit with no post-treatment cells never
  runs its bootstrap (overall ATT already NaN with a UserWarning), stores no surface, and
  aggregate('event_study') raises the absent-surface ValueError.
  • Materialization is FAIL-SOFT with a one-shot overall-only retry: an exception while
    building the surfaces, or a bootstrap attempt that raises / returns None with surfaces
    present, leaves them None, emits one UserWarning ("... could not be materialized ...")
    and re-runs the bootstrap overall-only (a seeded fit reproduces the first attempt's weight
    draws), so a plain bootstrapped fit's headline quintet and its success/abort outcome are
    unchanged by the removal. ContinuousDiD bootstrap failures still propagate as before.
  • Consequences on bootstrapped plain fits: the aggregators' own UserWarnings (ImputationDiD
    Prop-5 unidentified horizons, empty horizons, LOO single-unit auxiliaries; TwoStageDiD
    Stage-1/Stage-2 singular or rank-deficient fallbacks) and the per-target bootstrap-stats
    RuntimeWarnings now fire where a plain fit never emitted them; summary() prints the
    event-study/group tables and to_dataframe(level="event_study"|"group") returns frames
    instead of raising; DiagnosticReport / BusinessReport run heterogeneity (Imputation /
    TwoStage on the relayed group dict, ContinuousDiD on the event study) and parallel trends
    (Imputation / TwoStage) on the relayed surface instead of skipping with
    NotImplementedError; plot_event_study(results) plots them natively. Measured cost (199
    draws, 300 units x 12 periods): Imputation 0.01s -> 0.08s, TwoStage 0.01s -> 0.04s,
    Continuous unchanged at 0.05s; all three default to n_bootstrap=0.
• ImputationDiD's inert-df_convention warning predicate is now exact: the per-lead inference
  is reachable iff pretrends=True and the design is not replicate-weighted (M-127); the
  unreachable pretrends+replicate fit-time gate is deleted (the combination is still rejected
  post-fit by aggregate('event_study')).
• Runtime messages (plot_event_study, the results-class "not stored" hints,
  _reporting_helpers, DiagnosticReport docstrings), practitioner_next_steps advice for
  bootstrapped ImputationDiD / TwoStageDiD / ContinuousDiD fits, the llms*.txt guides, the
  ContinuousDiD API page and tutorial, the migration guide and the performance docs describe
  the relay instead of steering users to the removed fit-time route.

Removed

• The 3.9 rename shims are gone (ledger rows flipped to removed). Old keyword spellings
  now raise TypeError: DifferenceInDifferences.fit(time=) -> post= (M-030),
  permutation_test(time=) / leave_one_out_test(time=) -> post= (M-137, M-138),
  WooldridgeDiD.fit(cohort=) -> first_treat= (M-032), ChaisemartinDHaultfoeuille.fit(group=, controls=) -> unit=, covariates= (M-033, M-034), twowayfeweights(group=) -> unit=
  (M-097), the *_col spellings on HeterogeneousAdoptionDiD.fit, RegressionDiscontinuity.fit,
  RDPlot.fit, joint_pretrends_test, joint_homogeneity_test, did_had_pretest_workflow
  (M-035..M-042, M-088, M-089, M-098..M-112) and trim_weights(weight_col=) ->
  weights= (M-113), StackedDiD(clean_control=) -> control_group= (M-043),
  WooldridgeDiDResults.to_dataframe(aggregation=) -> level= (M-044) and
  summary(aggregation) (M-087; summary(alpha=None) is now the positional signature),
  ContinuousDiD(covariates=) -> fit(covariates=) (M-084). The "event" value spelling is
  rejected with ValueError on WooldridgeDiDResults.aggregate(type=), .to_dataframe(level=) and
  LPDiDResults.to_dataframe(level=) - use "event_study" (M-086, M-150, M-136).
• robust= removed from DifferenceInDifferences (and so TwoWayFixedEffects,
  MultiPeriodDiD), TripleDifference, HeterogeneousAdoptionDiD and LinearRegression
  (M-045..M-047, M-115); vcov_type= is the single selector. The legacy-CR1 remap that
  turned robust=False + clustering into hc1 went with it. Positional constructor slots after
  the removed parameter shift left by one (DiD slot 1, LinearRegression/TripleDifference slot 2,
  HeterogeneousAdoptionDiD slot 6); SyntheticDiD's inert lambda_reg/zeta (M-001, M-002,
  constructor slots 7-8) are removed too, shifting vcov_type and the conley_* parameters two
  slots left. Pass these by keyword.
• Results-field aliases removed: SyntheticDiDResults.placebo_effects (M-003; use
  variance_effects), RegressionDiscontinuityResults.treatment_col (M-094; takeup),
  StackedDiDResults.clean_control and StackedDiD.clean_control (M-095, M-043;
  control_group), ChaisemartinDHaultfoeuilleResults.groups (M-114; units),
  ChangesInChangesResults.estimator (M-143; method). Their serialized dual keys are gone from
  to_dict() (treatment_col, clean_control, estimator, Wooldridge's
  aggregation_weights["event"]); 3.x pickles still load (__setstate__ key migrations kept).
• The eight module-level wrapper functions are removed (M-070..M-077): imputation_did,
  two_stage_did, stacked_did, trop, synthetic_control, triple_difference,
  bacon_decompose, chaisemartin_dhaultfoeuille. Call the estimator classes. diff_diff.trop,
  diff_diff.synthetic_control, diff_diff.stacked_did and diff_diff.chaisemartin_dhaultfoeuille
  now name the submodules (as diff_diff.bacon always did).
• The dieted aliases are removed (M-132..M-134): CDiD -> ContinuousDiD, Stacked ->
  StackedDiD, Gardner -> TwoStageDiD; the module __getattr__ warning shim (M-135) is gone.
• The fit-time aggregate= / balance_e= keywords are gone on the estimators whose post-fit
  route already existed (ledger rows flipped to removed). fit(aggregate=) is removed from
  CallawaySantAnna (M-020), EfficientDiD (M-023), StackedDiD (M-024),
  ChaisemartinDHaultfoeuille (M-026) and HeterogeneousAdoptionDiD (M-027),
  fit(balance_e=) from CallawaySantAnna (M-117) and EfficientDiD (M-120), and
  aggregate= from did_had_pretest_workflow (M-139; the HADPretestReport.aggregate
  output field survives). The successor is post-fit results.aggregate(type, balance_e=); HAD
  selects its mode from the panel shape. With the keywords go StackedDiD's fit-time
  group/all/unknown-value ValueErrors (aggregate('group') already fails closed in the
  mixin) and the private _VALID_AGGREGATES sets in had.py / had_pretests.py
  (_infer_aggregate_mode is the sole selector). ImputationDiD, TwoStageDiD and
  ContinuousDiD lost their keywords in the same release once the bootstrapped-fit relay
  landed (the next bullet but one).
• CallawaySantAnna and EfficientDiD compute no event-study/group surface at fit. Their
  legacy event_study_effects / group_effects fields - and CS's cband_crit_value,
  event_study_vcov, event_study_vcov_index, event_study_df - are None after EVERY fit;
  the post-fit replay (aggregate('event_study'|'group', balance_e=)) is the only bootstrap
  route and is unchanged.
• The last fit-time aggregate= / balance_e= keywords are gone (ledger rows flipped to
  removed). fit(aggregate=) is removed from ImputationDiD (M-021), TwoStageDiD
  (M-022) and ContinuousDiD (M-025), fit(balance_e=) from ImputationDiD (M-118)
  and TwoStageDiD (M-119). With PR 3b's removals on CallawaySantAnna / EfficientDiD /
  StackedDiD / dCDH / HAD this completes the aggregate-postfit group: every adopter aggregates
  through post-fit results.aggregate(type, balance_e=) only; TripleDifference's
  staggered-mode keywords (M-140/M-141) were the last fit-time aggregate= at this step
  and are removed in the same release by the DDD container unification (Breaking Changes
  above). With the keywords go ContinuousDiD's
  no-underscore "eventstudy" spelling and its fit-time value validation, the dead analytic
  fit-time event-study inference arm in continuous_did.py, and ImputationDiD's unreachable
  pretrends+replicate fit-time gate.
• The StaggeredTripleDifference class and its SDDD alias are gone (M-013, M-064; 4.0
  cut, PR 4a). from diff_diff import StaggeredTripleDifference and SDDD raise ImportError,
  and the diff_diff.staggered_triple_diff module no longer exists. The staggered DDD estimator
  is TripleDifference().fit(..., unit=, time=, first_treat=, partition=) - the same engine
  (diff_diff/_staggered_triple_diff_engine.py), pinned against the committed oracle in
  tests/test_v4_merge_ddd.py. R's compact control_group spellings
  ("notyettreated"/"nevertreated") died with the class; the merged surface takes
  "not_yet_treated"/"never_treated".
• The QDiD class and the QDiDResults alias are gone (M-015, M-061). The quantile-DiD
  comparison estimator is ChangesInChanges(method="qdid") (same engine, identical numbers);
  the results container is ChangesInChangesResults for both methods.
• The two pyproject.toml pytest filterwarnings entries that silenced the removed classes'
  construction FutureWarnings.
• The MultiPeriodDiD class, its MultiPeriodDiDResults container, the PeriodEffect
  record and the EventStudy alias are gone (M-010, M-011, M-012, M-060; 4.0 cut,
  PR 4c-ii). from diff_diff import MultiPeriodDiD (and the other three names) raises
  ImportError. The estimator is TwoWayFixedEffects().fit(..., event_study=True, spec=, post_periods=, reference_period=), which returns the unified EventStudyResults surface
  natively: spec="pooled" is the removed class's design (bit-exact without unit=; with
  unit= the pooled spec auto-clusters at unit), spec="within" is its absorb=["unit"]
  design; post_periods= is required (the midpoint default died with the class) and the
  post-period average avg_att has no successor field. Pinned field for field against the
  pre-refactor capture tests/fixtures/v4_twfe_event_study_oracles.json.
• The pyproject.toml pytest filterwarnings entry that silenced the removed class's
  construction FutureWarning, the MultiPeriodDiD power-simulation profile
  (simulate_power has no event-study profile yet; TODO row), and the MultiPeriodDiDResults
  branches of DiagnosticReport / BusinessReport / practitioner_next_steps (both reports still
  reject EventStudyResults by design; the admission TODO row records where the logic lived).

Fixed

• DiagnosticReport kept every event-study horizon on fractional calendars. The parallel-trends and heterogeneity collectors keyed the derived aggregate('event_study') surface (and compared raw event_study_effects keys) through int(k), so on a half-step calendar (e = t - g in -1.5, -1.0, -0.5, ...) -1.5 merged into -1 and -0.5 into the treatment horizon 0: pre-periods silently left the PT family and a violated horizon could read as a clean verdict. Integer-valued horizons keep their int keys; fractional horizons stay float, compare at their real value against the anticipation-aware boundary, and align with the covariance index for the joint Wald. Affected every post-fit event-study producer routed through the report (CallawaySantAnna included), not only the staggered DDD route this release enables.

• run_all_placebo_tests and run_placebo_test(test_type="permutation" | "leave_one_out") no
  longer fail on a multi-period calendar column: through 3.x the wrappers forwarded time itself as
  the 0/1 post dummy, so DifferenceInDifferences.fit rejected it and the suite silently reported
  the two tests as error dicts (tutorial 04's placebo cell printed ERROR lines). The dummy is now
  derived from post_periods.

Security

• CI smoke-test lockfile refresh: bumped the transitive undici (7.29.0 -> 7.30.1)
  and source-map-js (1.2.1 -> 1.2.2) pins in .github/scripts/package-lock.json,
  clearing all seven open Dependabot alerts. CI-only (docs search-excerpt smoke test);
  nothing in the published wheel is affected.

Documentation

• 4.0-cut framing (the first PR of the 4.0 cut, at v3.13.0): docs/v4-design.md
  re-synced with the library as shipped since 3.9 - the estimator roster now counts
  LWDiD/DMLDiD/DurationDiD (27 -> 24 at 4.0, 23 estimators + Bacon in the diagnostic
  family), the diagnostic family names the TWFE weights containers and
  DurationDiDPretestResults, and section 9 carries the ten-PR cut sequence. M-080
  (auto-cluster policy) re-scoped to the measured roster: static TwoWayFixedEffects
  already auto-clusters at unit, SpilloverDiD is the one estimator whose default SE
  moves at 4.0, cluster=False becomes the uniform explicit off-switch, and the
  by-construction estimators keep no auto-cluster metadata by decision (new "Default
  clustering roster" section in docs/methodology/variance-conventions.md; REGISTRY
  Notes on SpilloverDiD, StackedDiD, ImputationDiD, TwoStageDiD, DMLDiD, DurationDiD).
  M-008 (DIFF_DIFF_SOLVE_OLS_FASTPATH default) decided: evaluated, kept off. New
  ledger row M-150 gives WooldridgeDiDResults.to_dataframe(level="event") its own
  locator (twin of M-136). The migration guide's auto-cluster section, appendix and
  orientation table follow the ledger (111 appendix rows).

• Tutorials 02/04/05/07/23 are rebuilt on TwoWayFixedEffects().fit(..., event_study=True)
  (spec="within": unit + period fixed effects, unit-clustered SEs) ahead of the
  MultiPeriodDiD removal (M-010, 4.0 cut PR 4c). Point estimates match the former
  MultiPeriodDiD fits on these balanced panels; the SEs change because the within spec
  clusters by unit. Tutorial 23's naive headline is the equal-weight average of the two
  post-period coefficients computed in the cell (ATT unchanged at -4.29; the SE is now the
  unit-clustered one), and tutorial 02's period-effects loop reads to_dataframe().

• The static TwoWayFixedEffects().fit(..., time=...) calls in tutorials 01/02/09 (the
  M-082 alias, misused there as the post dummy) are replaced: 01 passes post="post";
  02 and 09 report TwoWayFixedEffects().decompose(...).twfe_estimate, the unit + period
  fixed-effects coefficient their Goodman-Bacon sections decompose (the old cells printed
  treated x period artefacts: 0.6309 and 0.0001). Tutorial 02 cell 36's covariate draw is
  seeded. Tutorials 02 and 09 are re-executed on the Rust backend, which produced the
  committed outputs, so the seeded bootstrap cells are unchanged.

• Tutorials 08 and 16 describe StaggeredTripleDifference as the removed class (4.0 cut,
  PR 4a). diff_diff/guides/llms-full.txt's event-study snippet calls print(es.summary())
  (EventStudyResults has no print_summary).

• BaconDecomposition is re-homed into the diagnostics family ([M-090], item 9 of the 4.0 cut): it leaves the estimator catalogues - README ## Estimators, llms.txt ## Estimators, the llms-full.txt Estimators section and its alias line, the API-reference Estimators autosummary and module toctree, and the docs homepage Supported Estimators table - for their Diagnostics groupings. RDPlot moves from the API-reference Visualization autosummary into Diagnostics; RDPlot, RDDensityTest and the TWFE weight diagnostics leave the llms-full.txt Estimators section. Docs-only: the import path, the Bacon alias, the results class and every number are unchanged.

• The docs homepage gains a "Diagnostics & Sensitivity" table (BaconDecomposition, RDPlot, RDDensityTest, HonestDiD, PreTrendsPower, PowerAnalysis, DiagnosticReport, the parallel-trends / placebo / TWFE-weight functions and the HAD pretest workflow) next to the estimator table; the API-reference Estimators and Diagnostics intros describe the split.

• docs/methodology/REGISTRY.md archive framing for the classes removed at 4.0: the MultiPeriodDiD and StaggeredTripleDifference sections and the QDiD class-spelling subsection open with an archive Note, the table of contents marks them (archived) and states the archived-sections convention, the cross-reference tables name ChangesInChanges(method="qdid") instead of a standalone QDiD estimator, and the "20-estimator count" wording is gone. The migration guide gains the [M-090] bullet.

Testing

• tests/test_t23_spillover_tva_drift.py's naive_fit fixture mirrors the notebook's TWFE
  event-study headline cell, runs the fit with warnings escalated to errors, and pins the
  post-period-average identity and the unit-clustered vcov-block SE alongside the existing
  -4.29 endpoint.
• The MultiPeriodDiD-fixture suites (honest_did, pretrends, methodology pretrends/honest,
  visualization, Conley, survey, replicate weights, vcov_type, df_convention, the estimators
  event-study behaviour pins, the event-study surface roster, the DiagnosticReport /
  BusinessReport generic fixtures) now run on TwoWayFixedEffects().fit(..., event_study=True)
  (or a static TWFE fit for the report classes); MultiPeriodDiD's own tests, the MPD-only
  designs (multi-absorb, fixed_effects=, wild bootstrap, container-field assertions, the
  MPD-branch consumer tests) and the MPD-vs-TWFE parity oracles are unchanged until PR 4c-ii.
  New tests/test_twfe_event_study_decoupling.py, shared helpers
  tests/helpers/twfe_es_oracle_lanes.py, event_study_average.py and
  event_study_surface_builders.py; the variance-conventions inventory gains the two TWFE
  event-study rows; a TODO row tracks the one-way HC2-BM absorbed-FE R-parity lane.

Internal

• Codex reviewer gpt-6 campaign pre-registration (PR-1 of 2): tools/reviewer-eval
  pre-registers the gpt-5.6-sol -> gpt-6-astra evaluation (DECISION_RULE.md:
  symmetric regression gate over every ground-truth bug, FP gate, downgrade rule,
  verifier-confirmed leaks void a cell and contest the gate, stop conditions; the
  2026-07 rule is archived under campaigns/) and splits the grader-facing text into
  GRADING_RUBRIC.md (marker-based LEAK? flag whose list renders from
  reviewer_audit.LEAK_RULES). New engines: reviewer_verdict.py (gates, precedence,
  binding to the sealed artifacts, protocol guards, PROVISIONAL/DEFINITIVE) and
  reviewer_audit.py (post-unblinding session-log audit, informational and
  over-matching by design). run_eval.py gains audit and verdict, runs every
  codex call under an isolated CI-faithful CODEX_HOME, records seven-key git
  provenance per artifact and in the manifest, seals artifacts at run time
  (artifacts_sha), binds the blinded bundle (bundle_id, run_ids_sha,
  rubric/prompt sha) and refuses post-run stages on a dirty or moved tree; run
  holds an exclusive per-subdir lock for the whole attempt, records the OK cells of
  aborted attempts too and refuses a rerun that would re-dispatch any cell an
  earlier attempt recorded OK; review worktree leaves are salted opaque hashes. Harness CLI re-pinned to
  codex-cli 0.162.1 (arm E = gpt-6-astra @ xhigh added; CI codex-version pinned
  to the same, lockstep-tested with DEFAULT_MODEL, the plan-review pin, the
  control arm and the prose pins).

• openai_review.py: call_codex parses codex's stderr session events into
  session_commands (+ start/exec/other-event counts, truncation flag) on every
  exit path and prints a one-line summary; --timeout now also caps the codex
  backend's wall clock when passed explicitly (previously ignored there), 0
  means no timeout on both backends and negatives are rejected; PRICING
  corrected for gpt-5.6-* and extended with gpt-6-; gpt-6 counts as a
  reasoning family; blinding scrubs gpt-6/astra and Unicode-dash spellings.

• Codex reviewer swap gpt-5.6-sol -> gpt-6.1-sol (reviewer-eval 2026-10 campaigns, PR-2):
  two blinded, pre-registered 44-review campaigns on the frozen tools/reviewer-eval
  harness. Campaign 1 (B = gpt-5.6-sol vs E = gpt-6-astra, both @ xhigh, k=2): mechanical
  DEFINITIVE NO-GO on one regression (s4-cs-unbalanced-nan-outcome), accepted by the
  user; no astra swap (campaigns/2026-10-gpt6-VERDICT.md). Campaign 2 (B vs F =
  gpt-6.1-sol @ xhigh, k=2, fresh control sample, same rule re-issued for F):
  mechanical DEFINITIVE PARITY with zero rule-defined regressions, zero downgrades,
  0 vs 0 false positives, two favourable catch-status changes against one
  UNSTABLE-to-MISSED judgment flag, ~15% lower latency and half the input price; the
  user accepted PARITY as GO
  (campaigns/2026-10-gpt61sol-VERDICT.md). Both verdicts ship with their redacted
  grading records (campaigns/2026-10-gpt6*-grading/). Production pins move in lockstep
  to gpt-6.1-sol: the CI workflow model, openai_review.DEFAULT_MODEL and the
  plan-review CODEX_MODEL (carried over from the code-review campaign, not re-run
  through plan-review-eval; TODO.md row), with the plan-review CODEX_TIMEOUT_S raised
  2400 -> 3600 to equal the harness ceiling. Harness: arm F registered with the contrast
  (B, F), campaign 1's rule archived as campaigns/2026-10-gpt6-DECISION_RULE.md,
  run --configs defaults to the last registered shape, PRICING gains gpt-6.1-sol,
  arm F takes role=control (B retained as previous production), the candidate-exposure
  gate exempts the control arm so bare smoke keeps working after a swap, and the
  DEFERRED gpt-6-sol cost-fallback row is retired (codex-cli 0.162.1's ChatGPT-account
  login accepts the gpt-6 Sol tier; campaign 2 was that evaluation).

• Dependabot npm coverage: added an npm ecosystem entry for /.github/scripts
  so the smoke-test toolchain gets weekly version-update PRs (grouped minor/patch)
  instead of drifting into advisories between manual lockfile refreshes.

• Ledger cut window (docs/v4-deprecations.yaml header cut_window: "4.0",
  cut_base: "3.13.0"; spec section 11): rows scheduled at 4.0 may flip terminal
  ahead of the version bump so the removals land over many PRs while __version__
  stayed 3.13.0; tests/test_v4_matrix.py relaxes only the early-removal guard for the
  window version, refuses any other version bump while early-terminal rows exist, and
  keeps due enforcement unchanged. The id snapshot and ROW_COUNT_FLOOR now include
  M-148/M-149 (shipped by #838 without extending them) and M-150 (132 rows).
  M-091's roster records the four later-born Diagnostic containers; M-050's
  code_refs record that diff_diff/_deprecation.py survives 4.0 as pickle-migration
  infrastructure; M-045/M-115 name the legacy-CR1 remap branches their removal
  deletes. No maint/3.13 branch was cut (maintainer decision 2026-10-18); 3.13.x
  patches, if ever needed, come from the v3.13.0 tag.

• tests/test_v4_matrix.py removed-status probes: a function row that resolves to a submodule
  counts as absent; a param row absent from the explicit signature counts as removed despite
  **kwargs. Naming-guard allowlist extended for the kept body-local aliases, __setstate__ keys
  and shared vocabulary. The diff_diff._deprecation helpers survived this PR for the remaining
  M-031/M-082 and PR-3 shims (item 9b later reduced the module to its pickle-migration and
  required-arg helpers).

• 4.0 cut, PR 3a - tests-first migration to the post-fit aggregation containers. The
  test and benchmark suites now read event-study / group surfaces from
  results.aggregate("event_study") / results.aggregate("group") (the container API that
  has existed since 3.9) instead of the deprecated fit-time fit(aggregate=...) keyword and
  the legacy event_study_effects / group_effects fields, so the keyword removals in PRs 3b
  and 3c land against already-migrated suites. No library source changes and no ledger row
  flips in this PR; the fit-time keywords still work and still warn. Bootstrapped
  ImputationDiD / TwoStageDiD / ContinuousDiD sites stay on the fit-time route until PR 3c
  adds the post-fit relay for them.

• Committed per-backend aggregation oracles. tests/test_aggregate_contract.py no longer
  uses the fit-time route as its numerical reference: every analytic surface is pinned at
  rtol=1e-12 (confidence-interval bounds at 1e-9: scipy's t.ppf moved by up to 3.4e-11
  relative between 1.13 and 1.14+) and every bootstrapped CS / EfficientDiD replay at the
  replay-parity tolerances against tests/fixtures/v4_aggregate_oracles.json, captured on the unmodified
  tree at 8724fecd by tests/_capture_v4_aggregate_oracles.py under both weight backends
  (see tests/fixtures/README.md). New tests/helpers/legacy_surface.py (es_dict,
  group_dict, legacy_standin) gives consumer tests a faithful stand-in for the 3.x
  fit-time surface where their subject is the native (legacy-pickle) route.

• Benchmarks (benchmarks/python/benchmark_callaway.py, benchmark_imputation.py, the
  speed_review scripts) call aggregate() post-fit inside the timed region and emit the
  same JSON row shape the R refresh joins on.

• docs/v4-design.md section 9 records the three-way split of cut-sequence item 3 (3a
  tests-first, 3b mechanical removals, 3c the Imputation/TwoStage/ContinuousDiD removals +
  relay).

• tests/test_aggregate_contract.py: the Test*Shim classes, fit-time fixtures and
  inertness comparisons for the five estimators are retired in favour of Test*Removal
  pins (TypeError, signature, positional-slot binding); the committed oracle fixture is the
  numerical reference. tests/_capture_v4_aggregate_oracles.py is marked historical (it
  drives the removed route; runnable only at the recorded captured_at_head). Naming-guard
  allowlist entries for the llms.txt / llms-full.txt canonical post-fit aggregate
  vocabulary. Ledger row M-093 records that CS/EfficientDiD populate the legacy fields on
  no fit. warn_deprecated_kwarg loses its last library caller and is retained (spec section
  8 rule 11) with a direct unit pin.

• tests/test_aggregate_contract.py: the remaining Test*Shim classes, fit-time fixtures and
  fail-closed bootstrap pins are retired for Test*Removal pins (TypeError, signature,
  positional-slot binding) and relay pins - the relayed surfaces equal the stored legacy dicts
  row-for-row and match the per-backend bootstrap oracles - plus failure-injection pins for the
  fail-soft materialization and the storage rule. The estimator suites, DiagnosticReport,
  event_study_surface (bootstrapped Imputation / TwoStage / Continuous enrolled as genuine
  native producers) and practitioner tests take their relay pins; the module-level
  FutureWarning filters for the fit-time route are gone. Naming-guard allowlist entries for the
  guides' post-fit aggregate vocabulary are finalized. Ledger row M-093 gains the fit-side
  table of which producers populate the legacy fields; M-127 records the exact predicate.
  docs/v4-design.md section 6 carries the relay amendment and the HYBRIDS paragraph; section 9
  marks sub-item 3c shipped.

• The three StaggeredTripleDifference suites (tests/test_staggered_triple_diff.py,
  tests/test_methodology_staggered_triple_diff.py, tests/test_survey_staggered_ddd.py) are
  ported to TripleDifference's staggered mode (keyword fits; the R golden's compact
  control_group values are mapped), so the engine keeps its behavior, methodology and survey
  coverage. tests/test_v4_merge_ddd.py / tests/test_v4_merge_cic.py drop their old-vs-new
  parity lanes for Test*Removal pins; the two capture scripts are marked historical and
  make_2x2 lives in the CiC merge test.

• Ledger: M-013/M-015 -> removed, M-061/M-064/M-085 -> done; M-031 keeps
  shimmed with warning: null; M-080 drops the deleted module from its code_refs.
  Spec: cut-sequence item 4 split into 4a/4b/4c, the cross-row rule gains its alias clause.
  docs/doc-deps.yaml re-primes the staggered_triple_diff group on the results module.

• tests/test_v4_matrix.py no longer requires StaggeredTripleDifference/QDiD worked-example
  owners in the migration guide; the two autosummary stubs are deleted; REGISTRY keeps the
  ## StaggeredTripleDifference and ### QDiD sections as the engine's methodology record
  (archive framing lands with M-090).

• docs/v4-design.md cut-sequence item 4b marked shipped; item 4c lists the two remaining
  markdown-only MultiPeriodDiD mentions (06 cell 34, 16 cell 40). docs/doc-deps.yaml maps
  the TWFE module to tutorials 01/02/04/05/07/09/23. No ledger row flips.

• The TWFE event-study surface is built directly from the shared core's payload
  (4.0 cut, PR 4c-i; ledger rows M-010 / M-011): DifferenceInDifferences._fit_event_study_core
  returns a private _EventStudyCorePayload, TwoWayFixedEffects.fit(..., event_study=True)
  builds its EventStudyResults from it via results_base._from_event_study_payload, and only
  MultiPeriodDiD.fit still wraps the payload into MultiPeriodDiDResults. _from_mpd and the
  payload route share one calendar-surface builder, so they cannot drift; a committed oracle
  captured on the pre-refactor tree (tests/fixtures/v4_twfe_event_study_oracles.json, capture
  script tests/_capture_v4_merge_mpd_oracles.py) pins every field of the TWFE surface and of
  its MultiPeriodDiD twin across eleven variance lanes. No user-visible behavior changes.

• Spec item 4c is split into 4c-i (this PR) and 4c-ii (the removals); the 4c-ii bullet records
  the spec="within" benchmark-arm relabel and the M-016 view's record type (user decisions
  2026-10-24). Ledger M-010 / M-011 notes updated; no status flips.

• PR 4c-ii-b (docs/benchmarks half of the MultiPeriodDiD removal): REGISTRY / REPORTING /
  survey-theory / METHODOLOGY_REVIEW and the remaining user-doc prose stop describing the
  removed class as live; TODO rows 33/35/38/95/99 and three DEFERRED rows that governed the
  class are reworded or retired. The oracle fixture tests/fixtures/v4_twfe_event_study_oracles. json drops its mpd_twin records and the lane metadata loses its MPD twin fields (the TWFE
  records are byte-identical); the capture script documents the per-lane capture SHAs.

• Benchmarks: the live event-study arm is the new benchmarks/python/benchmark_twfe_event_study.py
  (spec="within", cluster="unit"; parity against R fixest::feols(... | unit) verified
  locally - ATT to 2e-12, SE to 0.0%, per-period effects correlation 1.000);
  benchmark_multiperiod.py and benchmarks/refresh_2026_07/ stay frozen because that harness
  runs inside pinned 3.5.3/3.7.0 venvs that predate event_study=True.

• DifferenceInDifferences._fit_event_study_core serves only the TWFE event-study mode
  (post_periods required by the core; include_treatment_main / cluster_override
  keyword-required); the private payload is pruned to the fields the surface builder reads.
  The clustered hc2_bm contrast batch keeps the former post-period-average column as the
  batch-relative Bell-McCaffrey DOF guard's anchor so per-period NaN masks stay bit-identical
  (REGISTRY Note; TODO row for the guard redesign). build_event_study_surface opens with an
  isinstance(EventStudyResults) passthrough.

• Ledger: M-010/M-011/M-012 -> removed, M-060/M-007/M-083 -> done, M-016
  -> shimmed on diff_diff:EventStudyResults.period_effects, M-082 keeps shimmed with
  warning: null. Spec item 4c-ii carries the SHIPPED note incl. the three recorded
  deviations (anchor column, new live benchmark script, DR/BR branch deletion).

• Tests: the MPD-only tests are deleted or re-fixtured onto cs_fit / the TWFE surface
  (tests/test_v4_merge_mpd.py carries the removal, M-016 and M-007 pins and the anchor
  pin); the oracle fixture gains two clustered hc2_bm lanes captured at d45d5316 before
  the core pruning (its mpd_twin records are pruned in PR 4c-ii-b); the clustered
  per-period CR2 DOF R/clubSandwich pin and the event-study covariate-collision guard are
  migrated to TWFE; variance-conventions twfe_es_* rows re-pinned (tail_df 5) and the table
  regenerated. The methodology/benchmark prose half and the oracle fixture's
  mpd_twin prune follow in PR 4c-ii-b.

• 4.0 cut, PR 5a-ii (the documentation half of cut-sequence item 5's reader migration,
  split from PR 5a because the combined diff exceeded the AI review input budget): every
  documented read of the 3.x overall_att / overall_se / overall_t_stat /
  overall_p_value / overall_conf_int names (and ContinuousDiD's overall_att_*
  spellings) on the nine M-050..M-058 containers - RTD pages, the bundled
  llms*.txt guides, tutorial code cells, the tutorial drift tests that mirror them
  (tutorial 32's code-cell hash pins re-pinned), benchmark scripts and the carousel
  generator - now reads the canonical att / se / t_stat / p_value / conf_int
  quintet. No user-visible behaviour change (the canonical names already resolve);
  committed notebook outputs are unchanged except tutorial 18's echoed guidance string.
  Bootstrap draw containers (bootstrap_results.overall_att_se) and R / golden-JSON
  interface keys are untouched.

• 4.0 cut, PR 5a (the reader half of cut-sequence item 5, the M-050..M-058 storage
  flip): every in-repo reader of the 3.x overall_att / overall_se / overall_t_stat /
  overall_p_value / overall_conf_int field names (and ContinuousDiD's overall_att_*
  spellings) on the nine flipping containers in SOURCE and TESTS - consumers, the
  containers' own summary() / to_dict() / __repr__ / aggregation bodies, and every
  test suite - now reads the canonical att / se /
  t_stat / p_value / conf_int quintet. No user-visible change: the canonical names
  already resolve as properties, numbers are bit-identical and nothing warns yet; the
  storage flip itself, the FutureWarning shims and the report-schema changes follow in
  PR 5b; the documentation / guides / tutorial / benchmark half of the reader migration
  is PR 5a-ii (split off because the combined diff exceeded the AI review input budget). power.py's _extract_staggered (a silent getattr fallback chain over the old
  names) is deleted - the staggered power profiles use _extract_simple; the
  practitioner_next_steps overall_att fallback is gone. Bootstrap draw containers
  (CSBootstrapResults.overall_att_se and siblings) and StaggeredTripleDiffResults
  (unified at cut item 6) are untouched.

• Ledger rows M-050..M-058 flip to shimmed with tests/test_v4_storage_flip.py as their test_ref; diff_diff._deprecation gains LEGACY_OVERALL_QUINTET and migrate_renamed_state_keys (the pickle-migration infrastructure); DMLDiDResults re-declares the five shims so the warning names the subclass (naming-guard allowlist reason reworded); _extract_scalar_headline's legacy alias row is deleted; tests/test_result_aliases.py pins the post-flip direction (SDDD kept in the 3.x direction); the API autosummary stubs are regenerated; the bootstrap draw containers' payload names are recorded as a REGISTRY Note (user decision 2026-10-26); spec item 5 carries its SHIPPED note.

• Ledger rows M-014, M-140, M-141 flip to removed with tests/test_v4_unify_ddd.py as their test_ref; diff_diff/staggered_triple_diff_results.py is renamed to diff_diff/triple_diff_results.py (hosting the unified container, re-exported from triple_diff.py); the staggered engine attaches the CallawaySantAnna aggregation kit (staggered.py's _build_aggregation_kit on the eligibility-zeroed precomputed copy, plus the ddd_df_survey_original bookkeeping key for the Eq. 4.14 df rule); report consumers dispatch on the design through _reporting_helpers.results_design / is_staggered_ddd (DR's _DESIGN_APPLICABILITY / _DESIGN_PT_METHOD); the practitioner handler gains a staggered branch; docs/api/_autosummary regenerated; the spec's section 4.2 / section 6 / item-6 SHIPPED notes and the REGISTRY ## TripleDifference post-fit Note record the mechanics and the weights='event_time' decision.

• Ledger rows M-004..M-006 and M-128..M-131 flip to done with tests/test_v4_inference_defaults.py as their test_ref; the spec's section 7 / item 7 SHIPPED notes and item 10 now record the deferred bulk re-execution of the nine tutorials whose clustered-default outputs moved (24 re-executed here if its outputs moved); the REGISTRY TwoWayFixedEffects clustered-CR1 inference-df note is relabelled a plain Note (the default now matches fixest) and the SunAbraham / ImputationDiD / StackedDiD / WooldridgeDiD / LPDiD tail-df notes read the 4.0 default; TestDfConvention, test_moderate_t_pins_cluster_df_convention and the per-estimator knob suites pin the G-1 default against explicit "residual" baselines (every default-vs-"cluster" comparison became default-vs-"residual" so none passes vacuously); test_ci_matches_r_twfe tightens to machine precision; the variance-conventions audit matrix keeps its residual-df rows as explicit "residual" arms (the K_full accounting pins) and gains eight G-1 _default_df twins (three in the fast set), with docs/methodology/variance-conventions.md regenerated; the committed TWFE event-study oracle lanes pooled_cluster_unit / within_cluster_unit and the ImputationDiD pretrends_all contract lanes name the "residual" convention they were captured under (metadata-only, no oracle number touched); the DEFERRED.md Version-gated (v4) row is retired.

• Ledger: M-080 and the new M-155 (shared cluster contract) -> done; new M-154 (SpilloverDiD df_convention, item 7c) -> planned; pinned by tests/test_v4_cluster_policy.py. Tutorial 23 section 6 shows the four-row SE table (CR1 at unit, HC1 via cluster=False, Conley lag 0 / lag 1) and was re-executed; tests/test_t23_spillover_tva_drift.py pins the new default and the table's source.

• Ledger: M-154 -> done, pinned by tests/test_v4_spillover_df_convention.py; the df ladder is the pure helper spillover._resolve_inference_df (survey df -> saturation -> utils.resolve_tail_df). The variance-conventions matrix gains the spillover_cluster_unit / spillover_cluster_unit_default_df rows; the REGISTRY df_convention knob roster, docs/methodology/variance-conventions.md, the migration guide, llms-full.txt and docs/api/spillover.rst describe the knob; tutorial 23 re-validated (it prints no p-values or CIs).

• results_base._from_relative_dict reads only the is_reference key and the named reference_period (the zero-count heuristic is gone; a keyed row with a finite nonzero effect raises); _deprecation.migrate_event_study_reference_sentinel is the single place the retired sentinel is still recognized, run by the carriers' __setstate__ (SpilloverDiDResults gained one; aggregate-mode frames untouched). Ledger: M-093 gains test_ref (tests/test_v4_sentinel_retirement.py) and stays planned until 8b (the to_dataframe(level="event_study") schema rewire on every results class, incl. the dCDH native frame's estimand 'ref' row). REGISTRY (SpilloverDiD, StackedDiD/ImputationDiD/TwoStageDiD, the plotting Notes), the migration guide, llms-full.txt and docs/api/spillover.rst describe the marking.

• The routing rule is the M-153 LWDiD precedent (aggregate(level).to_dataframe() on the AggregationMixin classes, build_event_study_surface(self).to_dataframe() elsewhere); results_base._ABSENT_SURFACE_HINTS gains a SunAbrahamResults entry. Ledger: M-093 flips to done with test_ref tests/test_v4_event_study_schema.py (the cross-producer schema-prefix / route-equality / error-contract pins; tests/test_v4_sentinel_retirement.py keeps the 8a pins) and a tenth amendment note; docs/v4-design.md section 5 now states the PREFIX rule and item 8 is complete. REGISTRY (CS, EfficientDiD, Imputation/TwoStage/Stacked relay and reference-row Notes, DDD, HAD, plus new to_dataframe-schema Notes for dCDH, SunAbraham, WooldridgeDiD, LPDiD, DurationDiD and SpilloverDiD), the migration guide, llms-full.txt, the practitioner guide, docs/api/wooldridge_etwfe.rst and docs/api/spillover.rst describe the frame; notebooks 16 (re-executed, outputs refreshed and prose reconciled - it leaves the TODO.md tutorial re-execution row), 19 (reads event_time/att/is_reference) and 14 (export prose) updated. TODO.md gains the level='group' relay row.

• Ledger row M-090 flips to done with tests/test_v4_diagnostic_family_docs.py as its test_ref; the test pins the estimators/diagnostics split on README, llms.txt, llms-full.txt, docs/index.rst, docs/api/index.rst and the REGISTRY archive framing, and runs in the docs-tests workflow (README-only diffs now trigger that lane). docs/doc-deps.yaml points the diagnostic modules at the new homepage table; the /docs-check classification table files Bacon as a diagnostics-class entry.

• Ledger rows M-016, M-050..M-058, M-151, M-152 flip to removed at 4.0 with
  deprecated_in: null (no release warned); M-148/M-149 flip to removed with
  deprecated_in: 3.13.0; new behavior rows M-156/M-157 (tests/test_diagnostics.py); the
  ledger header, spec ladder, section 5/8/9/11 notes and the phase-5 roster carry dated amendments.
  diff_diff._deprecation is reduced to the pickle-migration (LEGACY_OVERALL_QUINTET,
  migrate_renamed_state_keys, migrate_event_study_reference_sentinel) and required-arg
  (NOT_SUPPLIED, require_arg) helpers - deprecated_field_property, resolve_renamed_kwarg,
  warn_deprecated_kwarg and deprecated_kwarg_message are deleted with their last consumers.
  With the descriptors gone, assigning a legacy name to a results instance creates a shadow
  attribute instead of raising (the non-slotted dataclasses) - the naming guard's consumer sweep
  is the in-repo defense. Naming guard: the DMLDiDResults.overall_att surface entry and the
  _deprecation.py placebo_effects consumer entry are deleted, the two placebo wrappers join the
  rule-1 calendar time inventory, and the retired token families' remaining in-repo hits
  (body-local overall_att names, the kept __setstate__ key maps, methodology history) are
  allowlisted. API autosummary stubs regenerated; TODO.md's placebo-wrapper redesign row closed;
  tutorial 17 re-executed (its stored practitioner snippet echoed the old time='post' advice).