Fix CI review R1: boundary-horizon validation, placebo=False guard, docstrings

P0: _largest_consecutive_block now raises ValueError when boundary horizon
    (-1 or +1) is missing after finite-SE filtering instead of silently
    returning the full list (would produce wrong HonestDiD bounds).
P1: honest_did=True now rejects placebo=False early instead of silently
    returning honest_did_results=None with no warning.
P2: Added 3 regression tests (boundary -1 missing, boundary +1 missing,
    placebo=False + honest_did).
P3: Updated docstrings in honest_did.py (6 locations) and docs/llms.txt
    to include ChaisemartinDHaultfoeuilleResults alongside MultiPeriodDiD/CS.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: igerber

diff_diff/chaisemartin_dhaultfoeuille.py

@@ -957,6 +957,12 @@ def fit(
                 "Set L_max to compute DID^{pl}_l placebos that HonestDiD uses as "
                 "pre-period coefficients."
             )
+        if honest_did and not self.placebo:
+            raise ValueError(
+                "honest_did=True requires placebo computation. The estimator was "
+                "constructed with placebo=False. Use "
+                "ChaisemartinDHaultfoeuille(placebo=True) (the default)."
+            )
 
         # Pivot to (group x time) matrices for vectorized computations
         d_pivot = cell.pivot(index=group, columns=time, values="d_gt").reindex(

diff_diff/honest_did.py

@@ -559,7 +559,7 @@ def _extract_event_study_params(
 
     Parameters
     ----------
-    results : MultiPeriodDiDResults or CallawaySantAnnaResults
+    results : MultiPeriodDiDResults, CallawaySantAnnaResults, or ChaisemartinDHaultfoeuilleResults
         Estimation results with event study structure.
 
     Returns
@@ -886,8 +886,14 @@ def _largest_consecutive_block(times, boundary_val):
                     if not times:
                         return []
                     if boundary_val not in times:
-                        # No boundary value - take the block closest to it
-                        return times
+                        raise ValueError(
+                            f"HonestDiD requires horizon {boundary_val} in "
+                            f"the dCDH "
+                            f"{'placebo' if boundary_val < 0 else 'event study'}"
+                            f" surface, but it was removed by finite-SE "
+                            f"filtering. Retained horizons: {times}. Ensure "
+                            f"horizon {boundary_val} has a finite SE."
+                        )
                     # Expand outward from boundary_val
                     block = [boundary_val]
                     idx = times.index(boundary_val)
@@ -2197,7 +2203,7 @@ def fit(
 
         Parameters
         ----------
-        results : MultiPeriodDiDResults or CallawaySantAnnaResults
+        results : MultiPeriodDiDResults, CallawaySantAnnaResults, or ChaisemartinDHaultfoeuilleResults
             Results from event study estimation.
         M : float, optional
             Override the M parameter for this fit.
@@ -2515,7 +2521,7 @@ def sensitivity_analysis(
 
         Parameters
         ----------
-        results : MultiPeriodDiDResults or CallawaySantAnnaResults
+        results : MultiPeriodDiDResults, CallawaySantAnnaResults, or ChaisemartinDHaultfoeuilleResults
             Results from event study estimation.
         M_grid : list of float, optional
             Grid of M values to evaluate. If None, uses default grid
@@ -2614,7 +2620,7 @@ def breakdown_value(
 
         Parameters
         ----------
-        results : MultiPeriodDiDResults or CallawaySantAnnaResults
+        results : MultiPeriodDiDResults, CallawaySantAnnaResults, or ChaisemartinDHaultfoeuilleResults
             Results from event study estimation.
         tol : float
             Tolerance for binary search.
@@ -2669,7 +2675,7 @@ def compute_honest_did(
 
     Parameters
     ----------
-    results : MultiPeriodDiDResults or CallawaySantAnnaResults
+    results : MultiPeriodDiDResults, CallawaySantAnnaResults, or ChaisemartinDHaultfoeuilleResults
         Results from event study estimation.
     method : str
         Type of restriction ("smoothness", "relative_magnitude", "combined").
@@ -2705,7 +2711,7 @@ def sensitivity_plot(
 
     Parameters
     ----------
-    results : MultiPeriodDiDResults or CallawaySantAnnaResults
+    results : MultiPeriodDiDResults, CallawaySantAnnaResults, or ChaisemartinDHaultfoeuilleResults
         Results from event study estimation.
     method : str
         Type of restriction.

docs/llms.txt

@@ -20,7 +20,7 @@ diagnostic steps produces unreliable results.
 3. **Test parallel trends** — simple 2x2: `check_parallel_trends()`, `equivalence_test_trends()`; staggered: inspect CS event-study pre-period coefficients (generic PT tests are invalid for staggered designs). Insignificant pre-trends do NOT prove PT holds.
 4. **Choose estimator** — staggered adoption → CS/SA/BJS (NOT plain TWFE); few treated units → SDiD; factor confounding → TROP; simple 2x2 → DiD. Run `BaconDecomposition` to diagnose TWFE bias.
 5. **Estimate** — `estimator.fit(data, ...)`. Always print the cluster count first and choose inference method based on the result (cluster-robust if >= 50 clusters, wild bootstrap if fewer).
-6. **Sensitivity analysis** — `compute_honest_did(results)` for bounds under PT violations (MultiPeriodDiD/CS only), `run_all_placebo_tests()` for 2x2 falsification, specification comparisons for staggered designs.
+6. **Sensitivity analysis** — `compute_honest_did(results)` for bounds under PT violations (MultiPeriodDiD, CS, or dCDH), `run_all_placebo_tests()` for 2x2 falsification, specification comparisons for staggered designs.
 7. **Heterogeneity** — CS: `aggregate='group'`/`'event_study'`; SA: `results.event_study_effects`/`to_dataframe(level='cohort')`; subgroup re-estimation.
 8. **Robustness** — compare 2-3 estimators (CS vs SA vs BJS), MUST report with and without covariates (shows whether conditioning drives identification), present pre-trends and sensitivity bounds.
 

tests/test_chaisemartin_dhaultfoeuille.py

@@ -3275,6 +3275,15 @@ def test_honest_did_requires_lmax(self):
                 honest_did=True,
             )
 
+    def test_honest_did_rejects_placebo_false(self):
+        """honest_did=True with placebo=False raises ValueError."""
+        df = self._make_data()
+        with pytest.raises(ValueError, match="placebo=False"):
+            ChaisemartinDHaultfoeuille(seed=1, placebo=False).fit(
+                df, "outcome", "group", "period", "treatment",
+                L_max=2, honest_did=True,
+            )
+
     def test_honest_did_standalone(self):
         """compute_honest_did() on dCDH results matches honest_did=True."""
         from diff_diff.honest_did import compute_honest_did

tests/test_honest_did.py

@@ -1429,3 +1429,29 @@ def test_dcdh_empty_consecutive_block_raises(self):
             warnings.simplefilter("ignore")
             with pytest.raises(ValueError, match="No placebo horizons with finite SEs"):
                 compute_honest_did(results)
+
+    def test_dcdh_missing_boundary_minus1_raises(self):
+        """ValueError when horizon -1 has NaN SE (boundary required)."""
+        import warnings
+
+        results = self._fit_dcdh()
+        # Corrupt only horizon -1 SE; leave -2 intact
+        results.placebo_event_study[-1]["se"] = float("nan")
+
+        with warnings.catch_warnings():
+            warnings.simplefilter("ignore")
+            with pytest.raises(ValueError, match="requires horizon -1"):
+                compute_honest_did(results)
+
+    def test_dcdh_missing_boundary_plus1_raises(self):
+        """ValueError when horizon +1 has NaN SE (boundary required)."""
+        import warnings
+
+        results = self._fit_dcdh()
+        # Corrupt only horizon +1 SE
+        results.event_study_effects[1]["se"] = float("nan")
+
+        with warnings.catch_warnings():
+            warnings.simplefilter("ignore")
+            with pytest.raises(ValueError, match="requires horizon 1"):
+                compute_honest_did(results)