feat: add LWDiD estimator (Lee & Wooldridge 2025, 2026)

CHANGELOG.md

@@ -41,6 +41,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   (`benchmarks/python/coverage_lpdid_ra.py`, ≈0.95 coverage across cluster counts). Confirms
   the two previously-provisional REGISTRY notes (RA influence-function variance convention;
   pooled fixed-composition estimand) resolve in the library's favour with no estimator change.
+- **`LWDiD` (Lee & Wooldridge 2025, 2026 rolling-transformation DiD).** Unit-specific
+  demean/detrend converts panel to cross-section; supports staggered adoption with
+  never-treated / not-yet-treated control groups, RA/IPW/IPWRA estimation, and
+  cluster-robust inference. Alias `LW`.
 
 ## [3.6.0] - 2026-06-29
 

README.md

@@ -118,6 +118,7 @@ Full guide: `diff_diff.get_llm_guide("practitioner")`.
 - [StaggeredTripleDifference](https://diff-diff.readthedocs.io/en/stable/api/staggered.html#staggeredtripledifference) - Ortiz-Villavicencio & Sant'Anna (2025) staggered DDD with group-time ATT
 - [WooldridgeDiD](https://diff-diff.readthedocs.io/en/stable/api/wooldridge_etwfe.html) - Wooldridge (2023, 2025) ETWFE: saturated OLS, logit/Poisson QMLE (ASF-based ATT). Alias `ETWFE`.
 - [LPDiD](https://diff-diff.readthedocs.io/en/stable/api/lpdid.html) - Dube, Girardi, Jorda & Taylor (2025) Local Projections DiD: per-horizon long-difference event study on clean controls (no negative weighting), variance- or equally-weighted ATT, for absorbing or non-absorbing (reversible) treatment
+- [LWDiD](https://diff-diff.readthedocs.io/en/stable/api/lwdid.html) - Lee & Wooldridge (2025, 2026) rolling-transformation DiD: unit-specific demean/detrend converts panel to cross-section, staggered adoption, RA/IPW/IPWRA estimation. Alias `LW`.
 - [BaconDecomposition](https://diff-diff.readthedocs.io/en/stable/api/bacon.html) - Goodman-Bacon (2021) decomposition for diagnosing TWFE bias in staggered settings
 
 ## Diagnostics & Sensitivity

diff_diff/__init__.py

@@ -232,6 +232,43 @@
 from diff_diff.wooldridge_results import WooldridgeDiDResults
 from diff_diff.lpdid import LPDiD
 from diff_diff.lpdid_results import LPDiDResults
+from diff_diff.lwdid import LWDiD, lwdid, validate_staggered_data, is_never_treated
+from diff_diff.lwdid_results import LWDiDResults
+from diff_diff.lwdid_wild_bootstrap import wild_cluster_bootstrap, WildClusterBootstrapResult
+from diff_diff.lwdid_randomization import randomization_inference, RandomizationResult
+from diff_diff.lwdid_trend_diagnostics import (
+    test_parallel_trends, diagnose_heterogeneous_trends,
+    recommend_transformation, ParallelTrendsTestResult,
+)
+from diff_diff.lwdid_sensitivity import (
+    sensitivity_analysis, robustness_pre_periods,
+    sensitivity_no_anticipation, SensitivityResult,
+)
+from diff_diff.lwdid_visualization import (
+    plot_cohort_trends, plot_event_study as plot_lwdid_event_study,
+    plot_sensitivity as plot_lwdid_sensitivity,
+    plot_bootstrap_distribution,
+)
+from diff_diff.lwdid_exceptions import (
+    LWDIDError,
+    LWDIDWarning,
+    LWDIDInferenceError,
+    BootstrapConvergenceError,
+    RandomizationError,
+    DiagnosticError,
+    InsufficientPrePeriodsError,
+    NumericalWarning,
+    RandomizationWarning,
+    DiagnosticWarning,
+    SensitivityWarning,
+    VisualizationError,
+    VisualizationWarning,
+)
+from diff_diff.lwdid_clustering import (
+    diagnose_clustering, diagnose_clustering_from_data,
+    recommend_clustering_level,
+    ClusteringDiagnostics, ClusteringRecommendation,
+)
 from diff_diff.utils import (
     WildBootstrapResults,
     check_parallel_trends,
@@ -300,6 +337,7 @@
 ETWFE = WooldridgeDiD
 DCDH = ChaisemartinDHaultfoeuille
 HAD = HeterogeneousAdoptionDiD
+LW = LWDiD
 
 __version__ = "3.6.0"
 __all__ = [
@@ -386,6 +424,46 @@
     # LPDiD (Local Projections DiD)
     "LPDiD",
     "LPDiDResults",
+    # LWDiD (Lee & Wooldridge rolling transformation DiD)
+    "LWDiD",
+    "LWDiDResults",
+    "LW",
+    "wild_cluster_bootstrap",
+    "WildClusterBootstrapResult",
+    "randomization_inference",
+    "RandomizationResult",
+    "test_parallel_trends",
+    "diagnose_heterogeneous_trends",
+    "recommend_transformation",
+    "ParallelTrendsTestResult",
+    "sensitivity_analysis",
+    "robustness_pre_periods",
+    "sensitivity_no_anticipation",
+    "SensitivityResult",
+    "lwdid",
+    "plot_cohort_trends",
+    "plot_lwdid_event_study",
+    "plot_lwdid_sensitivity",
+    "plot_bootstrap_distribution",
+    # LWDiD exceptions
+    "LWDIDError",
+    "LWDIDWarning",
+    "LWDIDInferenceError",
+    "RandomizationError",
+    "DiagnosticError",
+    "NumericalWarning",
+    "DiagnosticWarning",
+    "SensitivityWarning",
+    "VisualizationError",
+    # LWDiD clustering diagnostics
+    "diagnose_clustering",
+    "diagnose_clustering_from_data",
+    "recommend_clustering_level",
+    "ClusteringDiagnostics",
+    "ClusteringRecommendation",
+    # LWDiD utility functions
+    "validate_staggered_data",
+    "is_never_treated",
     # Visualization
     "plot_bacon",
     "plot_event_study",

diff_diff/guides/llms.txt

@@ -70,6 +70,7 @@ Full practitioner guide: call `diff_diff.get_llm_guide("practitioner")`
 - [StaggeredTripleDifference](https://diff-diff.readthedocs.io/en/stable/api/staggered.html#staggeredtripledifference): Ortiz-Villavicencio & Sant'Anna (2025) staggered DDD with group-time ATT
 - [WooldridgeDiD](https://diff-diff.readthedocs.io/en/stable/api/wooldridge_etwfe.html): Wooldridge (2023, 2025) ETWFE — saturated OLS, logit/Poisson QMLE (ASF-based ATT). Alias: ETWFE
 - [LPDiD](https://diff-diff.readthedocs.io/en/stable/api/lpdid.html): Dube, Girardi, Jorda & Taylor (2025) Local Projections DiD: per-horizon long-difference event study on clean controls (no negative weighting); variance- or equally-weighted ATT, premean differencing, pooled pre/post, fast. Absorbing by default; non-absorbing (reversible) treatment via `non_absorbing="first_entry"` (Eq. 12) or `"effect_stabilization"` (Eq. 13, window `L`).
+- **LWDiD** (Lee & Wooldridge 2025, 2026): Rolling-transformation DiD — unit-specific demean/detrend converts panel to cross-section, supports staggered adoption with flexible control groups and estimation (RA/IPW/IPWRA). Alias: `LW`. [API](https://diff-diff.readthedocs.io/en/stable/api/lwdid.html)
 - [BaconDecomposition](https://diff-diff.readthedocs.io/en/stable/api/bacon.html): Goodman-Bacon (2021) decomposition for diagnosing TWFE bias in staggered settings
 
 ## Diagnostics and Sensitivity Analysis