diff-diff 1.2.0__tar.gz → 1.3.0__tar.gz

{diff_diff-1.2.0 → diff_diff-1.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: diff-diff
-Version: 1.2.0
+Version: 1.3.0
 Summary: A library for Difference-in-Differences causal inference analysis
 Author: diff-diff contributors
 License-Expression: MIT
@@ -106,6 +106,7 @@ Signif. codes: '***' 0.001, '**' 0.01, '*' 0.05, '.' 0.1
 - **Panel data support**: Two-way fixed effects estimator for panel designs
 - **Multi-period analysis**: Event-study style DiD with period-specific treatment effects
 - **Staggered adoption**: Callaway-Sant'Anna (2021) and Sun-Abraham (2021) estimators for heterogeneous treatment timing
+- **Triple Difference (DDD)**: Ortiz-Villavicencio & Sant'Anna (2025) estimators with proper covariate handling
 - **Synthetic DiD**: Combined DiD with synthetic control for improved robustness
 - **Event study plots**: Publication-ready visualization of treatment effects
 - **Parallel trends testing**: Multiple methods including equivalence tests
@@ -128,6 +129,8 @@ We provide Jupyter notebook tutorials in `docs/tutorials/`:
 | `04_parallel_trends.ipynb` | Testing parallel trends, equivalence tests, placebo tests, diagnostics |
 | `05_honest_did.ipynb` | Honest DiD sensitivity analysis, bounds, breakdown values, visualization |
 | `06_power_analysis.ipynb` | Power analysis, MDE, sample size calculations, simulation-based power |
+| `07_pretrends_power.ipynb` | Pre-trends power analysis (Roth 2022), MDV, power curves |
+| `08_triple_diff.ipynb` | Triple Difference (DDD) estimation with proper covariate handling |
 
 ## Data Preparation
 
@@ -901,6 +904,77 @@ print(f"Sun-Abraham ATT: {sa_results.overall_att:.3f}")
 # If results differ substantially, investigate heterogeneity
 ```
 
+### Triple Difference (DDD)
+
+Triple Difference (DDD) is used when treatment requires satisfying two criteria: belonging to a treated **group** AND being in an eligible **partition**. The `TripleDifference` class implements the methodology from Ortiz-Villavicencio & Sant'Anna (2025), which correctly handles covariate adjustment (unlike naive implementations).
+
+```python
+from diff_diff import TripleDifference, triple_difference
+
+# Basic usage
+ddd = TripleDifference(estimation_method='dr')  # doubly robust (recommended)
+results = ddd.fit(
+    data,
+    outcome='wages',
+    group='policy_state',       # 1=state enacted policy, 0=control state
+    partition='female',         # 1=women (affected by policy), 0=men
+    time='post'                 # 1=post-policy, 0=pre-policy
+)
+
+# View results
+results.print_summary()
+print(f"ATT: {results.att:.3f} (SE: {results.se:.3f})")
+
+# With covariates (properly incorporated, unlike naive DDD)
+results = ddd.fit(
+    data,
+    outcome='wages',
+    group='policy_state',
+    partition='female',
+    time='post',
+    covariates=['age', 'education', 'experience']
+)
+```
+
+**Estimation methods:**
+
+| Method | Description | When to use |
+|--------|-------------|-------------|
+| `"dr"` | Doubly robust | Recommended. Consistent if either outcome or propensity model is correct |
+| `"reg"` | Regression adjustment | Simple outcome regression with full interactions |
+| `"ipw"` | Inverse probability weighting | When propensity score model is well-specified |
+
+```python
+# Compare estimation methods
+for method in ['reg', 'ipw', 'dr']:
+    est = TripleDifference(estimation_method=method)
+    res = est.fit(data, outcome='y', group='g', partition='p', time='t')
+    print(f"{method}: ATT={res.att:.3f} (SE={res.se:.3f})")
+```
+
+**Convenience function:**
+
+```python
+# One-liner estimation
+results = triple_difference(
+    data,
+    outcome='wages',
+    group='policy_state',
+    partition='female',
+    time='post',
+    covariates=['age', 'education'],
+    estimation_method='dr'
+)
+```
+
+**Why use DDD instead of DiD?**
+
+DDD allows for violations of parallel trends that are:
+- Group-specific (e.g., economic shocks in treatment states)
+- Partition-specific (e.g., trends affecting women everywhere)
+
+As long as these biases are additive, DDD differences them out. The key assumption is that the *differential* trend between eligible and ineligible units would be the same across groups.
+
 ### Event Study Visualization
 
 Create publication-ready event study plots:
@@ -1696,6 +1770,60 @@ SunAbraham(
 | `print_summary(alpha)` | Print summary to stdout |
 | `to_dataframe(level)` | Convert to DataFrame ('event_study' or 'cohort') |
 
+### TripleDifference
+
+```python
+TripleDifference(
+    estimation_method='dr',   # 'dr' (doubly robust), 'reg', or 'ipw'
+    robust=True,              # Use HC1 robust standard errors
+    cluster=None,             # Column for cluster-robust SEs
+    alpha=0.05,               # Significance level for CIs
+    pscore_trim=0.01          # Propensity score trimming threshold
+)
+```
+
+**fit() Parameters:**
+
+| Parameter | Type | Description |
+|-----------|------|-------------|
+| `data` | DataFrame | Input data |
+| `outcome` | str | Outcome variable column name |
+| `group` | str | Group indicator column (0/1): 1=treated group |
+| `partition` | str | Partition/eligibility indicator column (0/1): 1=eligible |
+| `time` | str | Time indicator column (0/1): 1=post-treatment |
+| `covariates` | list | Covariate column names for adjustment |
+
+### TripleDifferenceResults
+
+**Attributes:**
+
+| Attribute | Description |
+|-----------|-------------|
+| `att` | Average Treatment effect on the Treated |
+| `se` | Standard error of ATT |
+| `t_stat` | T-statistic |
+| `p_value` | P-value for H0: ATT = 0 |
+| `conf_int` | Tuple of (lower, upper) confidence bounds |
+| `n_obs` | Total number of observations |
+| `n_treated_eligible` | Obs in treated group & eligible partition |
+| `n_treated_ineligible` | Obs in treated group & ineligible partition |
+| `n_control_eligible` | Obs in control group & eligible partition |
+| `n_control_ineligible` | Obs in control group & ineligible partition |
+| `estimation_method` | Method used ('dr', 'reg', or 'ipw') |
+| `group_means` | Dict of cell means for diagnostics |
+| `pscore_stats` | Propensity score statistics (IPW/DR only) |
+| `is_significant` | Boolean for significance at alpha |
+| `significance_stars` | String of significance stars |
+
+**Methods:**
+
+| Method | Description |
+|--------|-------------|
+| `summary(alpha)` | Get formatted summary string |
+| `print_summary(alpha)` | Print summary to stdout |
+| `to_dict()` | Convert to dictionary |
+| `to_dataframe()` | Convert to pandas DataFrame |
+
 ### HonestDiD
 
 ```python
@@ -2059,6 +2187,18 @@ This library implements methods from the following scholarly works:
 
 - **Arkhangelsky, D., Athey, S., Hirshberg, D. A., Imbens, G. W., & Wager, S. (2021).** "Synthetic Difference-in-Differences." *American Economic Review*, 111(12), 4088-4118. [https://doi.org/10.1257/aer.20190159](https://doi.org/10.1257/aer.20190159)
 
+### Triple Difference (DDD)
+
+- **Ortiz-Villavicencio, M., & Sant'Anna, P. H. C. (2025).** "Better Understanding Triple Differences Estimators." *Working Paper*. [https://arxiv.org/abs/2505.09942](https://arxiv.org/abs/2505.09942)
+
+  This paper shows that common DDD implementations (taking the difference between two DiDs, or applying three-way fixed effects regressions) are generally invalid when identification requires conditioning on covariates. The `TripleDifference` class implements their regression adjustment, inverse probability weighting, and doubly robust estimators.
+
+- **Gruber, J. (1994).** "The Incidence of Mandated Maternity Benefits." *American Economic Review*, 84(3), 622-641. [https://www.jstor.org/stable/2118071](https://www.jstor.org/stable/2118071)
+
+  Classic paper introducing the Triple Difference design for policy evaluation.
+
+- **Olden, A., & Møen, J. (2022).** "The Triple Difference Estimator." *The Econometrics Journal*, 25(3), 531-553. [https://doi.org/10.1093/ectj/utac010](https://doi.org/10.1093/ectj/utac010)
+
 ### Parallel Trends and Pre-Trend Testing
 
 - **Roth, J. (2022).** "Pretest with Caution: Event-Study Estimates after Testing for Parallel Trends." *American Economic Review: Insights*, 4(3), 305-322. [https://doi.org/10.1257/aeri.20210236](https://doi.org/10.1257/aeri.20210236)

{diff_diff-1.2.0 → diff_diff-1.3.0}/README.md

@@ -71,6 +71,7 @@ Signif. codes: '***' 0.001, '**' 0.01, '*' 0.05, '.' 0.1
 - **Panel data support**: Two-way fixed effects estimator for panel designs
 - **Multi-period analysis**: Event-study style DiD with period-specific treatment effects
 - **Staggered adoption**: Callaway-Sant'Anna (2021) and Sun-Abraham (2021) estimators for heterogeneous treatment timing
+- **Triple Difference (DDD)**: Ortiz-Villavicencio & Sant'Anna (2025) estimators with proper covariate handling
 - **Synthetic DiD**: Combined DiD with synthetic control for improved robustness
 - **Event study plots**: Publication-ready visualization of treatment effects
 - **Parallel trends testing**: Multiple methods including equivalence tests
@@ -93,6 +94,8 @@ We provide Jupyter notebook tutorials in `docs/tutorials/`:
 | `04_parallel_trends.ipynb` | Testing parallel trends, equivalence tests, placebo tests, diagnostics |
 | `05_honest_did.ipynb` | Honest DiD sensitivity analysis, bounds, breakdown values, visualization |
 | `06_power_analysis.ipynb` | Power analysis, MDE, sample size calculations, simulation-based power |
+| `07_pretrends_power.ipynb` | Pre-trends power analysis (Roth 2022), MDV, power curves |
+| `08_triple_diff.ipynb` | Triple Difference (DDD) estimation with proper covariate handling |
 
 ## Data Preparation
 
@@ -866,6 +869,77 @@ print(f"Sun-Abraham ATT: {sa_results.overall_att:.3f}")
 # If results differ substantially, investigate heterogeneity
 ```
 
+### Triple Difference (DDD)
+
+Triple Difference (DDD) is used when treatment requires satisfying two criteria: belonging to a treated **group** AND being in an eligible **partition**. The `TripleDifference` class implements the methodology from Ortiz-Villavicencio & Sant'Anna (2025), which correctly handles covariate adjustment (unlike naive implementations).
+
+```python
+from diff_diff import TripleDifference, triple_difference
+
+# Basic usage
+ddd = TripleDifference(estimation_method='dr')  # doubly robust (recommended)
+results = ddd.fit(
+    data,
+    outcome='wages',
+    group='policy_state',       # 1=state enacted policy, 0=control state
+    partition='female',         # 1=women (affected by policy), 0=men
+    time='post'                 # 1=post-policy, 0=pre-policy
+)
+
+# View results
+results.print_summary()
+print(f"ATT: {results.att:.3f} (SE: {results.se:.3f})")
+
+# With covariates (properly incorporated, unlike naive DDD)
+results = ddd.fit(
+    data,
+    outcome='wages',
+    group='policy_state',
+    partition='female',
+    time='post',
+    covariates=['age', 'education', 'experience']
+)
+```
+
+**Estimation methods:**
+
+| Method | Description | When to use |
+|--------|-------------|-------------|
+| `"dr"` | Doubly robust | Recommended. Consistent if either outcome or propensity model is correct |
+| `"reg"` | Regression adjustment | Simple outcome regression with full interactions |
+| `"ipw"` | Inverse probability weighting | When propensity score model is well-specified |
+
+```python
+# Compare estimation methods
+for method in ['reg', 'ipw', 'dr']:
+    est = TripleDifference(estimation_method=method)
+    res = est.fit(data, outcome='y', group='g', partition='p', time='t')
+    print(f"{method}: ATT={res.att:.3f} (SE={res.se:.3f})")
+```
+
+**Convenience function:**
+
+```python
+# One-liner estimation
+results = triple_difference(
+    data,
+    outcome='wages',
+    group='policy_state',
+    partition='female',
+    time='post',
+    covariates=['age', 'education'],
+    estimation_method='dr'
+)
+```
+
+**Why use DDD instead of DiD?**
+
+DDD allows for violations of parallel trends that are:
+- Group-specific (e.g., economic shocks in treatment states)
+- Partition-specific (e.g., trends affecting women everywhere)
+
+As long as these biases are additive, DDD differences them out. The key assumption is that the *differential* trend between eligible and ineligible units would be the same across groups.
+
 ### Event Study Visualization
 
 Create publication-ready event study plots:
@@ -1661,6 +1735,60 @@ SunAbraham(
 | `print_summary(alpha)` | Print summary to stdout |
 | `to_dataframe(level)` | Convert to DataFrame ('event_study' or 'cohort') |
 
+### TripleDifference
+
+```python
+TripleDifference(
+    estimation_method='dr',   # 'dr' (doubly robust), 'reg', or 'ipw'
+    robust=True,              # Use HC1 robust standard errors
+    cluster=None,             # Column for cluster-robust SEs
+    alpha=0.05,               # Significance level for CIs
+    pscore_trim=0.01          # Propensity score trimming threshold
+)
+```
+
+**fit() Parameters:**
+
+| Parameter | Type | Description |
+|-----------|------|-------------|
+| `data` | DataFrame | Input data |
+| `outcome` | str | Outcome variable column name |
+| `group` | str | Group indicator column (0/1): 1=treated group |
+| `partition` | str | Partition/eligibility indicator column (0/1): 1=eligible |
+| `time` | str | Time indicator column (0/1): 1=post-treatment |
+| `covariates` | list | Covariate column names for adjustment |
+
+### TripleDifferenceResults
+
+**Attributes:**
+
+| Attribute | Description |
+|-----------|-------------|
+| `att` | Average Treatment effect on the Treated |
+| `se` | Standard error of ATT |
+| `t_stat` | T-statistic |
+| `p_value` | P-value for H0: ATT = 0 |
+| `conf_int` | Tuple of (lower, upper) confidence bounds |
+| `n_obs` | Total number of observations |
+| `n_treated_eligible` | Obs in treated group & eligible partition |
+| `n_treated_ineligible` | Obs in treated group & ineligible partition |
+| `n_control_eligible` | Obs in control group & eligible partition |
+| `n_control_ineligible` | Obs in control group & ineligible partition |
+| `estimation_method` | Method used ('dr', 'reg', or 'ipw') |
+| `group_means` | Dict of cell means for diagnostics |
+| `pscore_stats` | Propensity score statistics (IPW/DR only) |
+| `is_significant` | Boolean for significance at alpha |
+| `significance_stars` | String of significance stars |
+
+**Methods:**
+
+| Method | Description |
+|--------|-------------|
+| `summary(alpha)` | Get formatted summary string |
+| `print_summary(alpha)` | Print summary to stdout |
+| `to_dict()` | Convert to dictionary |
+| `to_dataframe()` | Convert to pandas DataFrame |
+
 ### HonestDiD
 
 ```python
@@ -2024,6 +2152,18 @@ This library implements methods from the following scholarly works:
 
 - **Arkhangelsky, D., Athey, S., Hirshberg, D. A., Imbens, G. W., & Wager, S. (2021).** "Synthetic Difference-in-Differences." *American Economic Review*, 111(12), 4088-4118. [https://doi.org/10.1257/aer.20190159](https://doi.org/10.1257/aer.20190159)
 
+### Triple Difference (DDD)
+
+- **Ortiz-Villavicencio, M., & Sant'Anna, P. H. C. (2025).** "Better Understanding Triple Differences Estimators." *Working Paper*. [https://arxiv.org/abs/2505.09942](https://arxiv.org/abs/2505.09942)
+
+  This paper shows that common DDD implementations (taking the difference between two DiDs, or applying three-way fixed effects regressions) are generally invalid when identification requires conditioning on covariates. The `TripleDifference` class implements their regression adjustment, inverse probability weighting, and doubly robust estimators.
+
+- **Gruber, J. (1994).** "The Incidence of Mandated Maternity Benefits." *American Economic Review*, 84(3), 622-641. [https://www.jstor.org/stable/2118071](https://www.jstor.org/stable/2118071)
+
+  Classic paper introducing the Triple Difference design for policy evaluation.
+
+- **Olden, A., & Møen, J. (2022).** "The Triple Difference Estimator." *The Econometrics Journal*, 25(3), 531-553. [https://doi.org/10.1093/ectj/utac010](https://doi.org/10.1093/ectj/utac010)
+
 ### Parallel Trends and Pre-Trend Testing
 
 - **Roth, J. (2022).** "Pretest with Caution: Event-Study Estimates after Testing for Parallel Trends." *American Economic Review: Insights*, 4(3), 305-322. [https://doi.org/10.1257/aeri.20210236](https://doi.org/10.1257/aeri.20210236)

{diff_diff-1.2.0 → diff_diff-1.3.0}/diff_diff/__init__.py

@@ -81,6 +81,11 @@ from diff_diff.sun_abraham import (
     SunAbraham,
     SunAbrahamResults,
 )
+from diff_diff.triple_diff import (
+    TripleDifference,
+    TripleDifferenceResults,
+    triple_difference,
+)
 from diff_diff.utils import (
     WildBootstrapResults,
     check_parallel_trends,
@@ -98,7 +103,7 @@ from diff_diff.visualization import (
     plot_sensitivity,
 )
 
-__version__ = "1.2.0"
+__version__ = "1.2.1"
 __all__ = [
     # Estimators
     "DifferenceInDifferences",
@@ -107,6 +112,7 @@ __all__ = [
     "SyntheticDiD",
     "CallawaySantAnna",
     "SunAbraham",
+    "TripleDifference",
     # Bacon Decomposition
     "BaconDecomposition",
     "BaconDecompositionResults",
@@ -123,6 +129,8 @@ __all__ = [
     "GroupTimeEffect",
     "SunAbrahamResults",
     "SABootstrapResults",
+    "TripleDifferenceResults",
+    "triple_difference",
     # Visualization
     "plot_event_study",
     "plot_group_effects",

{diff_diff-1.2.0 → diff_diff-1.3.0}/diff_diff/staggered.py

@@ -1603,30 +1603,25 @@ class CallawaySantAnna:
                         weights * bootstrap_atts_gt[b, gt_indices]
                     )
 
-        # Compute bootstrap statistics
-        # ATT(g,t) statistics
+        # Compute bootstrap statistics for ATT(g,t)
         gt_ses = {}
         gt_cis = {}
         gt_p_values = {}
 
         for j, gt in enumerate(gt_pairs):
-            original_effect = original_atts[j]
-            boot_dist = bootstrap_atts_gt[:, j]
-
-            se = float(np.std(boot_dist, ddof=1))
-            ci = self._compute_percentile_ci(boot_dist, self.alpha)
-            p_value = self._compute_bootstrap_pvalue(original_effect, boot_dist)
-
+            se, ci, p_value = self._compute_effect_bootstrap_stats(
+                original_atts[j], bootstrap_atts_gt[:, j]
+            )
             gt_ses[gt] = se
             gt_cis[gt] = ci
             gt_p_values[gt] = p_value
 
-        # Overall ATT statistics
-        overall_se = float(np.std(bootstrap_overall, ddof=1))
-        overall_ci = self._compute_percentile_ci(bootstrap_overall, self.alpha)
-        overall_p_value = self._compute_bootstrap_pvalue(original_overall, bootstrap_overall)
+        # Compute bootstrap statistics for overall ATT
+        overall_se, overall_ci, overall_p_value = self._compute_effect_bootstrap_stats(
+            original_overall, bootstrap_overall
+        )
 
-        # Event study statistics
+        # Compute bootstrap statistics for event study effects
         event_study_ses = None
         event_study_cis = None
         event_study_p_values = None
@@ -1637,16 +1632,14 @@ class CallawaySantAnna:
             event_study_p_values = {}
 
             for e in rel_periods:
-                original_effect = event_study_info[e]['effect']
-                boot_dist = bootstrap_event_study[e]
-
-                event_study_ses[e] = float(np.std(boot_dist, ddof=1))
-                event_study_cis[e] = self._compute_percentile_ci(boot_dist, self.alpha)
-                event_study_p_values[e] = self._compute_bootstrap_pvalue(
-                    original_effect, boot_dist
+                se, ci, p_value = self._compute_effect_bootstrap_stats(
+                    event_study_info[e]['effect'], bootstrap_event_study[e]
                 )
+                event_study_ses[e] = se
+                event_study_cis[e] = ci
+                event_study_p_values[e] = p_value
 
-        # Group effect statistics
+        # Compute bootstrap statistics for group effects
         group_effect_ses = None
         group_effect_cis = None
         group_effect_p_values = None
@@ -1657,14 +1650,12 @@ class CallawaySantAnna:
             group_effect_p_values = {}
 
             for g in groups:
-                original_effect = group_agg_info[g]['effect']
-                boot_dist = bootstrap_group[g]
-
-                group_effect_ses[g] = float(np.std(boot_dist, ddof=1))
-                group_effect_cis[g] = self._compute_percentile_ci(boot_dist, self.alpha)
-                group_effect_p_values[g] = self._compute_bootstrap_pvalue(
-                    original_effect, boot_dist
+                se, ci, p_value = self._compute_effect_bootstrap_stats(
+                    group_agg_info[g]['effect'], bootstrap_group[g]
                 )
+                group_effect_ses[g] = se
+                group_effect_cis[g] = ci
+                group_effect_p_values[g] = p_value
 
         return CSBootstrapResults(
             n_bootstrap=self.n_bootstrap,
@@ -1817,6 +1808,35 @@ class CallawaySantAnna:
 
         return float(p_value)
 
+    def _compute_effect_bootstrap_stats(
+        self,
+        original_effect: float,
+        boot_dist: np.ndarray,
+    ) -> Tuple[float, Tuple[float, float], float]:
+        """
+        Compute bootstrap statistics for a single effect.
+
+        Parameters
+        ----------
+        original_effect : float
+            Original point estimate.
+        boot_dist : np.ndarray
+            Bootstrap distribution of the effect.
+
+        Returns
+        -------
+        se : float
+            Bootstrap standard error.
+        ci : Tuple[float, float]
+            Percentile confidence interval.
+        p_value : float
+            Bootstrap p-value.
+        """
+        se = float(np.std(boot_dist, ddof=1))
+        ci = self._compute_percentile_ci(boot_dist, self.alpha)
+        p_value = self._compute_bootstrap_pvalue(original_effect, boot_dist)
+        return se, ci, p_value
+
     def get_params(self) -> Dict[str, Any]:
         """Get estimator parameters (sklearn-compatible)."""
         return {