diffindiff 2.5.1__tar.gz → 2.5.2__tar.gz

{diffindiff-2.5.1 → diffindiff-2.5.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: diffindiff
-Version: 2.5.1
+Version: 2.5.2
 Summary: diffindiff: Python library for convenient Difference-in-Differences analyses
 Author: Thomas Wieland
 Author-email: geowieland@googlemail.com
@@ -22,14 +22,14 @@ Thomas Wieland [ORCID](https://orcid.org/0000-0001-5168-9846) [EMail](mailto:geo
 - 💻 GitHub Repository: [diffindiff_official](https://github.com/geowieland/diffindiff_official)
 - 📄 DOI (Zenodo): [10.5281/zenodo.18656820](https://doi.org/10.5281/zenodo.18656820)
 
-A research note featuring a case study that utilizes the diffindiff library is available on [arXiv](https://arxiv.org/abs/2605.21464).
+A case study that utilizes the diffindiff library is available on [arXiv](https://arxiv.org/abs/2605.21464).
 
 
 ## Citation
 
 If you use this software, please cite:
 
-Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.5.1) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
+Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.5.2) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
 
 
 ## Installation
@@ -53,6 +53,7 @@ pip install git+https://github.com/geowieland/diffindiff_official.git
   - Define custom treatment and control groups as well as treatment periods
   - Create ready-to-fit DiD data objects
   - Create predictive counterfactuals
+  - Create a synthetic control unit 
 - **DiD analysis**: 
   - Perform standard DiD analysis with pre-post data
   - Perform DiD analysis with two-way fixed effects models
@@ -66,6 +67,7 @@ pip install git+https://github.com/geowieland/diffindiff_official.git
     - Including after-treatment period
   - Perform Triple Difference (DDD) analysis
   - Perform DiD analysis with demeaned data
+  - Perform Synthetic DiD analysis
 - **Diagnosis tools and extensions of analyses**:
   - Add own counterfactuals or create counterfactuals based on machine learning or OLS regression models
   - Bonferroni correction for treatment effects
@@ -177,9 +179,11 @@ See the /tests directory for usage examples of most of the included functions.
 This software was developed without the use of AI-generated code. The Continue Agent in Microsoft Visual Studio Code using the GPT-5 mini model (by OpenAI) was used solely to assist in drafting and refining docstrings for documentation. The corresponding guidelines and constraints defined by the author are documented in `AGENTS-docstrings.md` in the [public GitHub repository](https://github.com/geowieland/diffindiff_official).
 
 
-## What's new (v2.5.1)
+## What's new (v2.5.2)
 
+- Extensions
+  - New functions DiffData.get_synthetic_control_weightings() and DiffData.get_synthetic_control_fitmetrics() for accessing control unit weightings and fit metrics of the synthetic control unit
 - Bugfixes
-  - In DiffData.create_treatment() and DiffData.create_data(), pre-post designs with after-treatment period (follow-up) now work correctly
-  - DiffModel.plot() nows supports pre-post designs with after-treatment period
-  - Some small stabilizations
+  - DiffModel.placebo() now checks control group size and returns warning when no. of control units is equal to 1, not allowing any placebo test
+- Other
+  - Some updates of documentation and README  

{diffindiff-2.5.1 → diffindiff-2.5.2}/README.md

@@ -14,14 +14,14 @@ Thomas Wieland [ORCID](https://orcid.org/0000-0001-5168-9846) [EMail](mailto:geo
 - 💻 GitHub Repository: [diffindiff_official](https://github.com/geowieland/diffindiff_official)
 - 📄 DOI (Zenodo): [10.5281/zenodo.18656820](https://doi.org/10.5281/zenodo.18656820)
 
-A research note featuring a case study that utilizes the diffindiff library is available on [arXiv](https://arxiv.org/abs/2605.21464).
+A case study that utilizes the diffindiff library is available on [arXiv](https://arxiv.org/abs/2605.21464).
 
 
 ## Citation
 
 If you use this software, please cite:
 
-Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.5.1) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
+Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.5.2) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
 
 
 ## Installation
@@ -45,6 +45,7 @@ pip install git+https://github.com/geowieland/diffindiff_official.git
   - Define custom treatment and control groups as well as treatment periods
   - Create ready-to-fit DiD data objects
   - Create predictive counterfactuals
+  - Create a synthetic control unit 
 - **DiD analysis**: 
   - Perform standard DiD analysis with pre-post data
   - Perform DiD analysis with two-way fixed effects models
@@ -58,6 +59,7 @@ pip install git+https://github.com/geowieland/diffindiff_official.git
     - Including after-treatment period
   - Perform Triple Difference (DDD) analysis
   - Perform DiD analysis with demeaned data
+  - Perform Synthetic DiD analysis
 - **Diagnosis tools and extensions of analyses**:
   - Add own counterfactuals or create counterfactuals based on machine learning or OLS regression models
   - Bonferroni correction for treatment effects
@@ -169,9 +171,11 @@ See the /tests directory for usage examples of most of the included functions.
 This software was developed without the use of AI-generated code. The Continue Agent in Microsoft Visual Studio Code using the GPT-5 mini model (by OpenAI) was used solely to assist in drafting and refining docstrings for documentation. The corresponding guidelines and constraints defined by the author are documented in `AGENTS-docstrings.md` in the [public GitHub repository](https://github.com/geowieland/diffindiff_official).
 
 
-## What's new (v2.5.1)
+## What's new (v2.5.2)
 
+- Extensions
+  - New functions DiffData.get_synthetic_control_weightings() and DiffData.get_synthetic_control_fitmetrics() for accessing control unit weightings and fit metrics of the synthetic control unit
 - Bugfixes
-  - In DiffData.create_treatment() and DiffData.create_data(), pre-post designs with after-treatment period (follow-up) now work correctly
-  - DiffModel.plot() nows supports pre-post designs with after-treatment period
-  - Some small stabilizations
+  - DiffModel.placebo() now checks control group size and returns warning when no. of control units is equal to 1, not allowing any placebo test
+- Other
+  - Some updates of documentation and README  

{diffindiff-2.5.1 → diffindiff-2.5.2}/diffindiff/config.py

@@ -4,15 +4,15 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     1.0.22
-# Last update: 2026-07-10 11:36
+# Version:     1.0.23
+# Last update: 2026-07-14 19:48
 # Copyright (c) 2025-2026 Thomas Wieland
 #-----------------------------------------------------------------------
 
 # Basic config:
 
 PACKAGE_NAME = "diffindiff"
-PACKAGE_VERSION = "2.5.1"
+PACKAGE_VERSION = "2.5.2"
 
 VERBOSE = False
 

{diffindiff-2.5.1 → diffindiff-2.5.2}/diffindiff/didanalysis.py

@@ -4,8 +4,8 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     2.4.2
-# Last update: 2026-07-10 11:30
+# Version:     2.4.3
+# Last update: 2026-07-14 19:49
 # Copyright (c) 2024-2026 Thomas Wieland
 #-----------------------------------------------------------------------
 
@@ -1422,10 +1422,15 @@ class DiffModel:
         model_data_c[time_col] = pd.to_datetime(model_data_c[time_col])
         model_data_c[unit_col] = model_data_c[unit_col].astype(str)
 
-        units_random_sample = model_data_c[unit_col].sample(
-            n = int(round(divide*control_group_N*resample, 0)), 
-            random_state = random_state
-            ).astype(str).tolist()
+        if control_group_N < 2:
+            print(f"WARNING: Model includes one control unit only: '{control_group[0]}'. Placebo analysis requires size of control group >1. Original model is returned.")
+            return self
+
+        else:
+            units_random_sample = model_data_c[unit_col].sample(
+                n = int(round(divide*control_group_N*resample, 0)), 
+                random_state = random_state
+                ).astype(str).tolist()
 
         model_data_c[TG_col_] = 0
         model_data_c.loc[(model_data_c[unit_col].isin(units_random_sample)), TG_col_] = 1

{diffindiff-2.5.1 → diffindiff-2.5.2}/diffindiff/didanalysis_helper.py

@@ -4,8 +4,8 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     1.2.2
-# Last update: 2026-07-02 20:18
+# Version:     1.2.3
+# Last update: 2026-07-14 19:44
 # Copyright (c) 2025-2026 Thomas Wieland
 #-----------------------------------------------------------------------
 

{diffindiff-2.5.1 → diffindiff-2.5.2}/diffindiff/diddata.py

@@ -4,8 +4,8 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     2.3.1
-# Last update: 2026-07-10 11:37
+# Version:     2.3.2
+# Last update: 2026-07-16 18:52
 # Copyright (c) 2024-2026 Thomas Wieland
 #-----------------------------------------------------------------------
 
@@ -815,7 +815,8 @@ class DiffData:
         unit_time_col_original,
         covariates,
         treatment_cols,
-        timestamp
+        timestamp,
+        synthetic_control
         ):
 
         """
@@ -840,6 +841,8 @@ class DiffData:
             Mapping of treatment columns metadata.
         timestamp : dict
             Timestamp metadata.
+        synthetic_control : dict
+            If included, data for a synthetic control unit, else None.
         Returns
         -------
         None
@@ -855,7 +858,8 @@ class DiffData:
             unit_time_col_original,
             covariates,
             treatment_cols,
-            timestamp
+            timestamp,
+            synthetic_control
             ]
 
     def get_did_modeldata_df (self):
@@ -1060,6 +1064,46 @@ class DiffData:
         """
 
         return self.data[7]
+    
+    def get_synthetic_control_weightings(self):
+
+        """
+        Return weightings of the synthetic control unit (if included in the DiffData object).
+
+        Returns
+        -------
+        pandas.DataFrame
+            Weighting by control unit.
+        
+        Examples
+        --------
+        >>> my_diffdata.get_synthetic_control_weightings()
+        """
+
+        if len(self.data[8]) > 0:
+            return self.data[8]["weights"]
+        else:
+            print("The DiffData object does not include a synthetic control unit.")
+            
+    def get_synthetic_control_fitmetrics(self):
+
+        """
+        Return fit metrics of the synthetic control unit (if included in the DiffData object).
+
+        Returns
+        -------
+        list
+            Fit metrics (dict) and data points with residuals (pandas.DataFrame).
+        
+        Examples
+        --------
+        >>> my_diffdata.get_synthetic_control_fitmetrics()
+        """
+
+        if len(self.data[8]) > 0:
+            return self.data[8]["fit_metrics"]
+        else:
+            print("The DiffData object does not include a synthetic control unit.")
 
     def add_covariates(
         self, 
@@ -2033,6 +2077,14 @@ class DiffData:
             verbose = False
             )
         
+        self.data[7] = helper.create_timestamp(function="add_synthetic")
+        
+        self.data[8] = {
+            "weights": weights_pd,
+            "fit_metrics": synthetic_fit_metrics,
+            "process_unit": process_unit
+        }
+        
         return self
  
     def summary(self):
@@ -2489,7 +2541,8 @@ def merge_data(
         unit_time_col_original,
         [],
         treatment_cols,
-        timestamp
+        timestamp,
+        {}
         )
 
     if verbose:

{diffindiff-2.5.1 → diffindiff-2.5.2}/diffindiff.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: diffindiff
-Version: 2.5.1
+Version: 2.5.2
 Summary: diffindiff: Python library for convenient Difference-in-Differences analyses
 Author: Thomas Wieland
 Author-email: geowieland@googlemail.com
@@ -22,14 +22,14 @@ Thomas Wieland [ORCID](https://orcid.org/0000-0001-5168-9846) [EMail](mailto:geo
 - 💻 GitHub Repository: [diffindiff_official](https://github.com/geowieland/diffindiff_official)
 - 📄 DOI (Zenodo): [10.5281/zenodo.18656820](https://doi.org/10.5281/zenodo.18656820)
 
-A research note featuring a case study that utilizes the diffindiff library is available on [arXiv](https://arxiv.org/abs/2605.21464).
+A case study that utilizes the diffindiff library is available on [arXiv](https://arxiv.org/abs/2605.21464).
 
 
 ## Citation
 
 If you use this software, please cite:
 
-Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.5.1) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
+Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.5.2) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
 
 
 ## Installation
@@ -53,6 +53,7 @@ pip install git+https://github.com/geowieland/diffindiff_official.git
   - Define custom treatment and control groups as well as treatment periods
   - Create ready-to-fit DiD data objects
   - Create predictive counterfactuals
+  - Create a synthetic control unit 
 - **DiD analysis**: 
   - Perform standard DiD analysis with pre-post data
   - Perform DiD analysis with two-way fixed effects models
@@ -66,6 +67,7 @@ pip install git+https://github.com/geowieland/diffindiff_official.git
     - Including after-treatment period
   - Perform Triple Difference (DDD) analysis
   - Perform DiD analysis with demeaned data
+  - Perform Synthetic DiD analysis
 - **Diagnosis tools and extensions of analyses**:
   - Add own counterfactuals or create counterfactuals based on machine learning or OLS regression models
   - Bonferroni correction for treatment effects
@@ -177,9 +179,11 @@ See the /tests directory for usage examples of most of the included functions.
 This software was developed without the use of AI-generated code. The Continue Agent in Microsoft Visual Studio Code using the GPT-5 mini model (by OpenAI) was used solely to assist in drafting and refining docstrings for documentation. The corresponding guidelines and constraints defined by the author are documented in `AGENTS-docstrings.md` in the [public GitHub repository](https://github.com/geowieland/diffindiff_official).
 
 
-## What's new (v2.5.1)
+## What's new (v2.5.2)
 
+- Extensions
+  - New functions DiffData.get_synthetic_control_weightings() and DiffData.get_synthetic_control_fitmetrics() for accessing control unit weightings and fit metrics of the synthetic control unit
 - Bugfixes
-  - In DiffData.create_treatment() and DiffData.create_data(), pre-post designs with after-treatment period (follow-up) now work correctly
-  - DiffModel.plot() nows supports pre-post designs with after-treatment period
-  - Some small stabilizations
+  - DiffModel.placebo() now checks control group size and returns warning when no. of control units is equal to 1, not allowing any placebo test
+- Other
+  - Some updates of documentation and README  

{diffindiff-2.5.1 → diffindiff-2.5.2}/setup.py

@@ -7,7 +7,7 @@ def read_README():
     
 setup(
     name='diffindiff',
-    version='2.5.1',
+    version='2.5.2',
     description='diffindiff: Python library for convenient Difference-in-Differences analyses',
     packages=find_packages(include=["diffindiff", "diffindiff.tests"]),
     include_package_data=True,