Address PR #207 review feedback: fix misleading control unit label and panel=FALSE golden values

- Rename summary label "Control units:" → "Never-treated units:" in CallawaySantAnnaResults
  to avoid confusion when not_yet_treated control group has zero never-treated units
- Clarify n_control_units docstring to note it excludes not-yet-treated dynamic controls
- Remove panel=FALSE from scenarios 4 and 5 in R golden value generator so golden
  values match the Python panel estimator (previously overstated validation)
- Regenerate csdid_golden_values.json

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

Author: igerber

benchmarks/R/generate_csdid_test_values.R

@@ -169,13 +169,13 @@ data4 <- build_sim_dataset(sp4)
 
 res4 <- att_gt(yname = "Y", xformla = ~X, data = data4, tname = "period",
                idname = "id", gname = "G", est_method = "reg",
-               bstrap = FALSE, cband = FALSE, panel = FALSE)
+               bstrap = FALSE, cband = FALSE)
 agg4_dynamic <- aggte(res4, type = "dynamic", bstrap = FALSE, cband = FALSE)
 
 scenarios$dynamic_effects <- list(
   data = export_data(data4),
   params = list(est_method = "reg", control_group = "nevertreated",
-                xformla = "~X", base_period = "varying", panel = FALSE),
+                xformla = "~X", base_period = "varying"),
   results = list(
     group_time = extract_gt(res4),
     dynamic = extract_agg(agg4_dynamic)
@@ -198,15 +198,15 @@ data5 <- subset(data5, period %in% keep.periods)
 
 res5 <- att_gt(yname = "Y", xformla = ~X, data = data5, tname = "period",
                idname = "id", gname = "G", est_method = "reg",
-               bstrap = FALSE, cband = FALSE, panel = FALSE)
+               bstrap = FALSE, cband = FALSE)
 agg5_dynamic <- aggte(res5, type = "dynamic", bstrap = FALSE, cband = FALSE)
 agg5_balance <- aggte(res5, type = "dynamic", balance_e = 0,
                       bstrap = FALSE, cband = FALSE)
 
 scenarios$non_consecutive_periods <- list(
   data = export_data(data5),
   params = list(est_method = "reg", control_group = "nevertreated",
-                xformla = "~X", base_period = "varying", panel = FALSE),
+                xformla = "~X", base_period = "varying"),
   results = list(
     group_time = extract_gt(res5),
     dynamic = extract_agg(agg5_dynamic),

benchmarks/data/csdid_golden_values.json

@@ -155,21 +155,20 @@
         "est_method": "reg",
         "control_group": "nevertreated",
         "xformla": "~X",
-        "base_period": "varying",
-        "panel": false
+        "base_period": "varying"
       },
       "results": {
         "group_time": {
           "group": [2, 2, 2, 3, 3, 3, 4, 4, 4],
           "time": [2, 3, 4, 2, 3, 4, 2, 3, 4],
-          "att": [1.1619345587, 1.9806271422, 3.1291417943, -0.11164281617, 0.8368911508, 2.1879948095, -0.10395702362, -0.00099808793357, 0.92702821049],
-          "se": [0.37188664974, 0.41090066647, 0.46080728316, 0.45290896601, 0.55834155256, 0.61699201548, 0.40706432128, 0.49486177347, 0.59510165336]
+          "att": [1.1619345587, 1.9806271422, 3.1291417943, -0.11164281617, 0.8368911508, 2.1879948095, -0.10395702362, -0.00099808793362, 0.92702821049],
+          "se": [0.19192101496, 0.21040775677, 0.20357402241, 0.22247872369, 0.21167719772, 0.22518098115, 0.20616133048, 0.20504607933, 0.21443054603]
         },
         "dynamic": {
           "overall_att": 2.060851975,
-          "overall_se": 0.27407895206,
+          "overall_se": 0.12749075561,
           "att": [-0.10395702362, -0.044806660534, 0.97874816264, 2.074665968, 3.1291417943],
-          "se": [0.40706432128, 0.32905421365, 0.29393486639, 0.35874990004, 0.46080728316],
+          "se": [0.20616133048, 0.13857947004, 0.10343596176, 0.15426829989, 0.20357402241],
           "egt": [-2, -1, 0, 1, 2]
         }
       }
@@ -187,28 +186,27 @@
         "est_method": "reg",
         "control_group": "nevertreated",
         "xformla": "~X",
-        "base_period": "varying",
-        "panel": false
+        "base_period": "varying"
       },
       "results": {
         "group_time": {
           "group": [2, 2, 2, 5, 5, 5, 7, 7, 7],
           "time": [2, 5, 7, 2, 5, 7, 2, 5, 7],
           "att": [0.99169165575, 3.9498224909, 5.5125753853, 0.0060151208735, 0.76227117049, 2.8066993777, -0.18717828366, -0.29655357309, 0.79303584629],
-          "se": [0.68280178613, 0.84102367846, 1.050254212, 0.67004481625, 0.85687016966, 0.95893447545, 0.62746934691, 0.79798902953, 1.2013519687]
+          "se": [0.28333148004, 0.27124277852, 0.2649377156, 0.27000914872, 0.26082165208, 0.24574534739, 0.29030567969, 0.28842055307, 0.27221422428]
         },
         "dynamic": {
           "overall_att": 3.2780871625,
-          "overall_se": 0.54064463051,
+          "overall_se": 0.15246801609,
           "att": [-0.18717828366, 0.0060151208735, -0.29655357309, 0.8432513962, 2.8066993777, 3.9498224909, 5.5125753853],
-          "se": [0.62746934691, 0.67004481625, 0.79798902953, 0.52758456271, 0.95893447545, 0.84102367846, 1.050254212],
+          "se": [0.29030567969, 0.27000914872, 0.28842055307, 0.12222233584, 0.24574534739, 0.27124277852, 0.2649377156],
           "egt": [-5, -3, -2, 0, 2, 3, 5]
         },
         "dynamic_balance_e0": {
           "overall_att": 0.8432513962,
-          "overall_se": 0.52758456271,
+          "overall_se": 0.12222233584,
           "att": [-0.18717828366, 0.0060151208735, -0.29655357309, 0.8432513962],
-          "se": [0.62746934691, 0.67004481625, 0.79798902953, 0.52758456271],
+          "se": [0.29030567969, 0.27000914872, 0.28842055307, 0.12222233584],
           "egt": [-5, -3, -2, 0]
         }
       }

diff_diff/staggered_results.py

@@ -88,7 +88,7 @@ class CallawaySantAnnaResults:
     n_treated_units : int
         Number of ever-treated units.
     n_control_units : int
-        Number of never-treated units.
+        Number of never-treated units (excludes not-yet-treated dynamic controls).
     event_study_effects : dict, optional
         Effects aggregated by relative time (event study).
     group_effects : dict, optional
@@ -152,7 +152,7 @@ def summary(self, alpha: Optional[float] = None) -> str:
             "",
             f"{'Total observations:':<30} {self.n_obs:>10}",
             f"{'Treated units:':<30} {self.n_treated_units:>10}",
-            f"{'Control units:':<30} {self.n_control_units:>10}",
+            f"{'Never-treated units:':<30} {self.n_control_units:>10}",
             f"{'Treatment cohorts:':<30} {len(self.groups):>10}",
             f"{'Time periods:':<30} {len(self.time_periods):>10}",
             f"{'Control group:':<30} {self.control_group:>10}",