Codex re-review round 4: propagate vcov_type / conley_lag_cutoff to to_dict

P1 (Maintainability): the prior commit added `conley_lag_cutoff` to the
result dataclasses and summary(), but `DiDResults.to_dict()` and
`MultiPeriodDiDResults.to_dict()` still omitted it. Downstream programmatic
consumers (notebooks, adapters, pipelines) that serialize results to dicts
couldn't tell which Conley variant produced the SEs.

Fix: both `to_dict()` methods now include `vcov_type`, `cluster_name`, and
`conley_lag_cutoff` when set. Conditional emission preserves the existing
behavior for non-conley / non-cluster fits (no new keys appear in the
serialized dict for unrelated estimators).

Regression: `test_twfe_conley_to_dict_carries_lag_cutoff` and
`test_multi_period_did_conley_to_dict_carries_lag_cutoff` fit a TWFE +
MPD Conley panel and assert `to_dict()` exposes the expected fields.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Author: igerber

diff_diff/results.py

@@ -284,6 +284,14 @@ def to_dict(self) -> Dict[str, Any]:
             "r_squared": self.r_squared,
             "inference_method": self.inference_method,
         }
+        # Variance-family metadata: included only when set, so existing
+        # dict consumers see no new keys for non-conley / non-cluster fits.
+        if self.vcov_type is not None:
+            result["vcov_type"] = self.vcov_type
+        if self.cluster_name is not None:
+            result["cluster_name"] = self.cluster_name
+        if self.conley_lag_cutoff is not None:
+            result["conley_lag_cutoff"] = self.conley_lag_cutoff
         if self.n_bootstrap is not None:
             result["n_bootstrap"] = self.n_bootstrap
         if self.n_clusters is not None:
@@ -717,6 +725,14 @@ def to_dict(self) -> Dict[str, Any]:
             "r_squared": self.r_squared,
             "reference_period": self.reference_period,
         }
+        # Variance-family metadata: included only when set, so existing
+        # dict consumers see no new keys for non-conley / non-cluster fits.
+        if self.vcov_type is not None:
+            result["vcov_type"] = self.vcov_type
+        if self.cluster_name is not None:
+            result["cluster_name"] = self.cluster_name
+        if self.conley_lag_cutoff is not None:
+            result["conley_lag_cutoff"] = self.conley_lag_cutoff
 
         # Add period-specific effects
         for period, pe in self.period_effects.items():

tests/test_conley_vcov.py

@@ -1243,6 +1243,51 @@ def test_multi_period_did_conley_with_datetime64_time(self):
         se_dt = np.sqrt(np.diag(res_dt.vcov))
         np.testing.assert_allclose(se_dt, se_int, atol=1e-10)
 
+    def test_multi_period_did_conley_to_dict_carries_lag_cutoff(self):
+        """Closes Codex re-review round 4 P1 (Maintainability) on MPD:
+        serialized `to_dict()` must include `vcov_type` and
+        `conley_lag_cutoff` so downstream programmatic consumers can tell
+        which Conley variant produced the SEs."""
+        import pandas as pd
+
+        from diff_diff import MultiPeriodDiD
+
+        rng = np.random.default_rng(seed=41)
+        n_units = 10
+        rows = []
+        for u in range(n_units):
+            lat = rng.uniform(-30, 30)
+            lon = rng.uniform(-100, 100)
+            for t in range(3):
+                rows.append(
+                    {
+                        "unit": u,
+                        "time": t,
+                        "y": rng.normal(),
+                        "treated": int(u < 5),
+                        "lat": lat,
+                        "lon": lon,
+                    }
+                )
+        df_mp = pd.DataFrame(rows)
+        res = MultiPeriodDiD(
+            vcov_type="conley",
+            conley_coords=("lat", "lon"),
+            conley_cutoff_km=2000.0,
+            conley_lag_cutoff=2,
+        ).fit(
+            df_mp,
+            outcome="y",
+            treatment="treated",
+            time="time",
+            post_periods=[1, 2],
+            unit="unit",
+            reference_period=0,
+        )
+        d = res.to_dict()
+        assert d["vcov_type"] == "conley"
+        assert d["conley_lag_cutoff"] == 2
+
     def test_multi_period_did_conley_missing_coords_raises(self):
         """MPD + vcov_type='conley' without conley_coords raises a clean
         ValueError instead of a raw TypeError on `self.conley_coords[0]`.
@@ -1421,6 +1466,24 @@ def test_twfe_conley_summary_emits_conley_label(self, panel):
         # The result dataclass also carries the lag for programmatic access.
         assert res.conley_lag_cutoff == 1
 
+    def test_twfe_conley_to_dict_carries_lag_cutoff(self, panel):
+        """Closes Codex re-review round 4 P1 (Maintainability): the
+        serialized `to_dict()` must include `vcov_type` and
+        `conley_lag_cutoff` so downstream programmatic consumers (notebooks,
+        adapters, pipelines) can tell which Conley variant produced the SEs
+        without re-deriving from the summary string."""
+        from diff_diff import TwoWayFixedEffects
+
+        res = TwoWayFixedEffects(
+            vcov_type="conley",
+            conley_coords=("lat", "lon"),
+            conley_cutoff_km=2000.0,
+            conley_lag_cutoff=1,
+        ).fit(panel, outcome="y", treatment="treated", time="time", unit="unit")
+        d = res.to_dict()
+        assert d["vcov_type"] == "conley"
+        assert d["conley_lag_cutoff"] == 1
+
     def test_twfe_conley_non_numeric_time_fails(self, panel):
         """TWFE's `_treatment_post = treated * time` design step requires
         numeric `time`. Non-numeric labels (datetime64, pd.Period, strings)