Survey Phase 7: CS IPW/DR covariates, repeated cross-sections, HonestDiD survey variance

ROADMAP.md

@@ -8,19 +8,42 @@ For past changes and release history, see [CHANGELOG.md](CHANGELOG.md).
 
 ## Current Status
 
-diff-diff v2.6.0 is a **production-ready** DiD library with feature parity with R's `did` + `HonestDiD` + `synthdid` ecosystem for core DiD analysis:
+diff-diff v2.7.5 is a **production-ready** DiD library with feature parity with R's `did` + `HonestDiD` + `synthdid` ecosystem for core DiD analysis, plus **unique survey support** — design-based variance estimation (Taylor linearization, replicate weights) integrated across all estimators. No R or Python package offers this combination:
 
 - **Core estimators**: Basic DiD, TWFE, MultiPeriod, Callaway-Sant'Anna, Sun-Abraham, Borusyak-Jaravel-Spiess Imputation, Synthetic DiD, Triple Difference (DDD), TROP, Two-Stage DiD (Gardner 2022), Stacked DiD (Wing et al. 2024), Continuous DiD (Callaway, Goodman-Bacon & Sant'Anna 2024)
 - **Valid inference**: Robust SEs, cluster SEs, wild bootstrap, multiplier bootstrap, placebo-based variance
 - **Assumption diagnostics**: Parallel trends tests, placebo tests, Goodman-Bacon decomposition
 - **Sensitivity analysis**: Honest DiD (Rambachan-Roth), Pre-trends power analysis (Roth 2022)
 - **Study design**: Power analysis tools
 - **Data utilities**: Real-world datasets (Card-Krueger, Castle Doctrine, Divorce Laws, MPDTA), DGP functions for all supported designs
+- **Survey support**: Full `SurveyDesign` with strata, PSU, FPC, weight types, replicate weights (BRR/Fay/JK1/JKn), Taylor linearization, DEFF diagnostics, subpopulation analysis — integrated across all estimators (see [survey-roadmap.md](docs/survey-roadmap.md))
 - **Performance**: Optional Rust backend for accelerated computation; faster than R at scale (see [CHANGELOG.md](CHANGELOG.md) for benchmarks)
 
 ---
 
-## Near-Term Enhancements (v2.7)
+## Near-Term Enhancements (v2.8)
+
+### Survey Phase 7: Completing the Survey Story
+
+Close the remaining gaps for practitioners using major population surveys
+(ACS, CPS, BRFSS, MEPS). See [survey-roadmap.md](docs/survey-roadmap.md) for
+full details.
+
+- **CS Covariates + IPW/DR + Survey** *(High priority)*: Implement DRDID
+  nuisance IF corrections under survey weights. Currently the recommended DR
+  method raises `NotImplementedError` with covariates + survey. This is the
+  most commonly needed path in applied work (Medicaid expansion, minimum wage).
+- **Repeated Cross-Sections** *(High priority)*: `panel=False` support for
+  CallawaySantAnna, enabling analysis of surveys that don't track units over
+  time (BRFSS, ACS annual, CPS monthly). Uses cross-sectional DRDID
+  (Sant'Anna & Zhao 2020, Section 4).
+- **Survey-Aware DiD Tutorial** *(High priority)*: Jupyter notebook
+  demonstrating the full workflow with realistic survey data. diff-diff is
+  the only package (R or Python) with design-based variance for modern DiD
+  — this makes that capability discoverable.
+- **HonestDiD + Survey Variance** *(Medium priority)*: Pass survey vcov
+  (TSL or replicate) into sensitivity analysis instead of cluster-robust vcov,
+  so sensitivity bounds respect the same variance structure as main estimates.
 
 ### Staggered Triple Difference (DDD)
 
@@ -32,12 +55,6 @@ Extend the existing `TripleDifference` estimator to handle staggered adoption se
 
 **Reference**: [Ortiz-Villavicencio & Sant'Anna (2025)](https://arxiv.org/abs/2505.09942). *Working Paper*. R package: `triplediff`.
 
-### Enhanced Visualization
-
-- Synthetic control weight visualization (bar chart of unit weights)
-- Treatment adoption "staircase" plot for staggered designs
-- Interactive plots with plotly backend option
-
 ---
 
 ## Medium-Term Enhancements

TODO.md

@@ -54,11 +54,17 @@ Deferred items from PR reviews that were not addressed before merge.
 |-------|----------|----|----------|
 | ImputationDiD dense `(A0'A0).toarray()` scales O((U+T+K)^2), OOM risk on large panels | `imputation.py` | #141 | Medium (deferred — only triggers when sparse solver fails) |
 | Multi-absorb weighted demeaning needs iterative alternating projections for N > 1 absorbed FE with survey weights; unweighted multi-absorb also uses single-pass (pre-existing, exact only for balanced panels) | `estimators.py` | #218 | Medium |
-| CallawaySantAnna survey + covariates + IPW/DR: DRDID panel nuisance-estimation IF corrections not implemented. Currently gated with NotImplementedError. Regression method with covariates works. | `staggered.py` | #233 | Medium — tracked in Survey Phase 7a |
-| EfficientDiD `control_group="last_cohort"` trims at `last_g - anticipation` but REGISTRY says `t >= last_g`. With `anticipation=0` (default) these are identical. Needs design decision for `anticipation>0`. | `efficient_did.py` | #230 | Low |
-| TripleDifference power: `generate_ddd_data` is a fixed 2×2×2 cross-sectional DGP — no multi-period or unbalanced-group support. | `prep_dgp.py`, `power.py` | #208 | Low |
-| Survey design resolution/collapse patterns inconsistent across panel estimators — extract shared helpers for panel-to-unit collapse, post-filter re-resolution, metadata recomputation | `continuous_did.py`, `efficient_did.py`, `stacked_did.py` | #226 | Low |
-| TROP: `fit()` and `_fit_global()` share ~150 lines of near-identical data setup. Extract shared helpers to eliminate cross-file sync risk. | `trop.py`, `trop_global.py`, `trop_local.py` | — | Low |
+| Replicate-weight survey df — **Resolved**. `df_survey = rank(replicate_weights) - 1` matching R's `survey::degf()`. For IF paths, `n_valid - 1` when dropped replicates reduce effective count. | `survey.py` | #238 | Resolved |
+| CallawaySantAnna survey: strata/PSU/FPC — **Resolved**. Aggregated SEs (overall, event study, group) use `compute_survey_if_variance()`. Bootstrap uses PSU-level multiplier weights. | `staggered.py` | #237 | Resolved |
+| CallawaySantAnna survey + covariates + IPW/DR — **Resolved**. DRDID panel nuisance IF corrections (PS + OR) implemented for both survey and non-survey DR paths (Phase 7a). IPW path unblocked. | `staggered.py` | #233 | Resolved |
+| SyntheticDiD/TROP survey: strata/PSU/FPC — **Resolved**. Rao-Wu rescaled bootstrap implemented for both. TROP uses cross-classified pseudo-strata. Rust TROP remains pweight-only (Python fallback for full design). | `synthetic_did.py`, `trop.py` | — | Resolved |
+| EfficientDiD hausman_pretest() clustered covariance stale `n_cl` — **Resolved**. Recompute `n_cl` and remap indices after `row_finite` filtering via `np.unique(return_inverse=True)`. | `efficient_did.py` | #230 | Resolved |
+| EfficientDiD `control_group="last_cohort"` trims at `last_g - anticipation` but REGISTRY says `t >= last_g`. With `anticipation=0` (default) these are identical. With `anticipation>0`, code is arguably more conservative (excludes anticipation-contaminated periods). Either align REGISTRY with code or change code to `t < last_g` — needs design decision. | `efficient_did.py` | #230 | Low |
+| TripleDifference power: `generate_ddd_data` is a fixed 2×2×2 cross-sectional DGP — no multi-period or unbalanced-group support. Add a `generate_ddd_panel_data` for panel DDD power analysis. | `prep_dgp.py`, `power.py` | #208 | Low |
+| ContinuousDiD event-study aggregation anticipation filter — **Resolved**. `_aggregate_event_study()` now filters `e < -anticipation` when `anticipation > 0`, matching CallawaySantAnna behavior. Bootstrap paths also filtered. | `continuous_did.py` | #226 | Resolved |
+| Survey design resolution/collapse patterns are inconsistent across panel estimators — ContinuousDiD rebuilds unit-level design in SE code, EfficientDiD builds once in fit(), StackedDiD re-resolves on stacked data; extract shared helpers for panel-to-unit collapse, post-filter re-resolution, and metadata recomputation | `continuous_did.py`, `efficient_did.py`, `stacked_did.py` | #226 | Low |
+| Survey metadata formatting dedup — **Resolved**. Extracted `_format_survey_block()` helper in `results.py`, replaced 13 occurrences across 11 files. | `results.py` + 10 results files | — | Resolved |
+| TROP: `fit()` and `_fit_global()` share ~150 lines of near-identical data setup (panel pivoting, absorbing-state validation, first-treatment detection, effective rank, NaN warnings). Both bootstrap methods also duplicate the stratified resampling loop. Extract shared helpers to eliminate cross-file sync risk. | `trop.py`, `trop_global.py`, `trop_local.py` | — | Low |
 
 #### Performance
 
@@ -161,8 +167,8 @@ Features in R's `did` package that block porting additional tests:
 
 | Feature | R tests blocked | Priority | Status |
 |---------|----------------|----------|--------|
-| Repeated cross-sections (`panel=FALSE`) | ~7 tests in test-att_gt.R + test-user_bug_fixes.R | High | Planned — Survey Phase 7b |
-| Sampling/population weights | 7 tests incl. all JEL replication | Medium | Mostly resolved (Phases 1-6); CS IPW/DR + covariates + survey in Phase 7a |
+| Repeated cross-sections (`panel=FALSE`) | ~7 tests in test-att_gt.R + test-user_bug_fixes.R | High | **Resolved** — Phase 7b: `panel=False` on CallawaySantAnna |
+| Sampling/population weights | 7 tests incl. all JEL replication | Medium | **Resolved** (Phases 1-6 + 7a: CS IPW/DR + covariates + survey) |
 | Calendar time aggregation | 1 test in test-att_gt.R | Low | |
 
 ---