diffindiff 2.3.0__tar.gz → 2.3.2__tar.gz

{diffindiff-2.3.0 → diffindiff-2.3.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: diffindiff
-Version: 2.3.0
+Version: 2.3.2
 Summary: diffindiff: Python library for convenient Difference-in-Differences analyses
 Author: Thomas Wieland
 Author-email: geowieland@googlemail.com
@@ -27,7 +27,7 @@ Thomas Wieland [ORCID](https://orcid.org/0000-0001-5168-9846) [EMail](mailto:geo
 
 If you use this software, please cite:
 
-Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.3.0) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
+Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.3.2) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
 
 
 ## Installation
@@ -52,10 +52,11 @@ pip install git+https://github.com/geowieland/diffindiff_official.git
   - Create ready-to-fit DiD data objects
   - Create predictive counterfactuals
 - **DiD analysis**: 
-  - Perfom standard DiD analysis with pre-post data
-  - Perfom DiD analysis with two-way fixed effects models
+  - Perform standard DiD analysis with pre-post data
+  - Perform DiD analysis with two-way fixed effects models
   - Simultaneous and/or staggered adoption are supported
   - Single or multiple treatments are supported
+  - Binary or continuous treatments are supported
   - Model extensions for DiD analysis:
     - Group- or individual-specific treatment effects
     - Group- or individual-specific time trends
@@ -66,10 +67,10 @@ pip install git+https://github.com/geowieland/diffindiff_official.git
   - Add own counterfactuals or create counterfactuals based on machine learning or OLS regression models
   - Bonferroni correction for treatment effects
   - Placebo test
-  - Test for control conditions
-  - Test for type of adoption
-  - Test whether the panel dataset is balanced
-  - Test for parallel trend assumption
+  - Test for control conditions (automatically within analysis or stand-alone)
+  - Test for type of adoption (automatically within analysis or stand-alone)
+  - Test whether the panel dataset is balanced (automatically within analysis or stand-alone)
+  - Test for parallel trend assumption (automatically within analysis or stand-alone)
 - **Visualization**:
   - Plot observed and expected time course of treatment and control group
   - Plot expected time course of treatment group and counterfactual
@@ -169,19 +170,10 @@ See the /tests directory for usage examples of most of the included functions.
 
 ## AI Usage Statement
 
-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.md` in the [public GitHub repository](https://github.com/geowieland/diffindiff_official).
+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 (v2.3.0)
+## What's new (v2.3.2)
 
-- General
-  - Full documentation (docstring in NumPy style) of all classes, methods, and functions
-  - Extended information in README
-- Extensions
-  - Timestamps for all methods creating or changing DiffGroups, DiffTreatment, DiffData and DiffModel objects
-  - Hyperparameters and other config information in didtools.model_wrapper() are now saved
-  - All summary methods return self
-- Bugfixes:
-  - Correct check of input lists in didanalysis_helper functions create_spillover() and extract_model_results() 
-  - Correct check of input lists in DiffData methods add_covariates() and analysis()
-  - Correct check of input lists in didanalysis functions did_analysis() and ddd_analysis()
+- Extensions:
+  - Re-transform log-transformed outcomes in DiffModel.plot() via parameter 'retransform_log_outcome'

{diffindiff-2.3.0 → diffindiff-2.3.2}/README.md

@@ -19,7 +19,7 @@ Thomas Wieland [ORCID](https://orcid.org/0000-0001-5168-9846) [EMail](mailto:geo
 
 If you use this software, please cite:
 
-Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.3.0) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
+Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.3.2) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
 
 
 ## Installation
@@ -44,10 +44,11 @@ pip install git+https://github.com/geowieland/diffindiff_official.git
   - Create ready-to-fit DiD data objects
   - Create predictive counterfactuals
 - **DiD analysis**: 
-  - Perfom standard DiD analysis with pre-post data
-  - Perfom DiD analysis with two-way fixed effects models
+  - Perform standard DiD analysis with pre-post data
+  - Perform DiD analysis with two-way fixed effects models
   - Simultaneous and/or staggered adoption are supported
   - Single or multiple treatments are supported
+  - Binary or continuous treatments are supported
   - Model extensions for DiD analysis:
     - Group- or individual-specific treatment effects
     - Group- or individual-specific time trends
@@ -58,10 +59,10 @@ pip install git+https://github.com/geowieland/diffindiff_official.git
   - Add own counterfactuals or create counterfactuals based on machine learning or OLS regression models
   - Bonferroni correction for treatment effects
   - Placebo test
-  - Test for control conditions
-  - Test for type of adoption
-  - Test whether the panel dataset is balanced
-  - Test for parallel trend assumption
+  - Test for control conditions (automatically within analysis or stand-alone)
+  - Test for type of adoption (automatically within analysis or stand-alone)
+  - Test whether the panel dataset is balanced (automatically within analysis or stand-alone)
+  - Test for parallel trend assumption (automatically within analysis or stand-alone)
 - **Visualization**:
   - Plot observed and expected time course of treatment and control group
   - Plot expected time course of treatment group and counterfactual
@@ -161,19 +162,10 @@ See the /tests directory for usage examples of most of the included functions.
 
 ## AI Usage Statement
 
-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.md` in the [public GitHub repository](https://github.com/geowieland/diffindiff_official).
+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 (v2.3.0)
+## What's new (v2.3.2)
 
-- General
-  - Full documentation (docstring in NumPy style) of all classes, methods, and functions
-  - Extended information in README
-- Extensions
-  - Timestamps for all methods creating or changing DiffGroups, DiffTreatment, DiffData and DiffModel objects
-  - Hyperparameters and other config information in didtools.model_wrapper() are now saved
-  - All summary methods return self
-- Bugfixes:
-  - Correct check of input lists in didanalysis_helper functions create_spillover() and extract_model_results() 
-  - Correct check of input lists in DiffData methods add_covariates() and analysis()
-  - Correct check of input lists in didanalysis functions did_analysis() and ddd_analysis()
+- Extensions:
+  - Re-transform log-transformed outcomes in DiffModel.plot() via parameter 'retransform_log_outcome'

{diffindiff-2.3.0 → diffindiff-2.3.2}/diffindiff/config.py

@@ -4,15 +4,15 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     1.0.8
-# Last update: 2026-02-27 21:55
+# Version:     1.0.10
+# Last update: 2026-03-05 21:27
 # Copyright (c) 2025-2026 Thomas Wieland
 #-----------------------------------------------------------------------
 
 # Basic config:
 
 PACKAGE_NAME = "diffindiff"
-PACKAGE_VERSION = "2.3.0"
+PACKAGE_VERSION = "2.3.2"
 
 VERBOSE = False
 

{diffindiff-2.3.0 → diffindiff-2.3.2}/diffindiff/didanalysis.py

@@ -4,8 +4,8 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     2.3.0
-# Last update: 2026-03-01 11:23
+# Version:     2.3.2
+# Last update: 2026-03-06 21:26
 # Copyright (c) 2024-2026 Thomas Wieland
 #-----------------------------------------------------------------------
 
@@ -186,12 +186,23 @@ class DiffModel:
         study_period_end = study_period_end.date()
         study_period_N = model_data[time_col].nunique()
         
-        if len(model_data[model_data[treatment] == 1]) > 0:
-            treatment_period_start = pd.to_datetime(min(model_data[model_data[treatment] == 1][time_col]))
-            treatment_period_end = pd.to_datetime(max(model_data[model_data[treatment] == 1][time_col]))
-            treatment_period_N = model_data.loc[model_data[treatment] == 1, time_col].nunique()
+        if config.ACCEPT_CONTINUOUS_TREATMENTS:
+            
+            if len(model_data[model_data[treatment] > 0]) > 0:
+                treatment_period_start = pd.to_datetime(min(model_data[model_data[treatment] > 0][time_col]))
+                treatment_period_end = pd.to_datetime(max(model_data[model_data[treatment] > 0][time_col]))
+                treatment_period_N = model_data.loc[model_data[treatment] > 0, time_col].nunique()
+            else:
+                treatment_period_N = 0
+        
         else:
-            treatment_period_N = 0
+        
+            if len(model_data[model_data[treatment] == 1]) > 0:
+                treatment_period_start = pd.to_datetime(min(model_data[model_data[treatment] == 1][time_col]))
+                treatment_period_end = pd.to_datetime(max(model_data[model_data[treatment] == 1][time_col]))
+                treatment_period_N = model_data.loc[model_data[treatment] == 1, time_col].nunique()
+            else:
+                treatment_period_N = 0
             
         after_treatment_period_start = None
         after_treatment_period_end = None
@@ -200,7 +211,10 @@ class DiffModel:
             after_treatment_period_start = treatment_period_end+pd.Timedelta(days=1)
             after_treatment_period_start = pd.to_datetime(after_treatment_period_start)
             after_treatment_period_end = pd.to_datetime(study_period_end)
-            after_treatment_period_N = model_data.loc[model_data[after_treatment_col] == 1, time_col].nunique()            
+            if config.ACCEPT_CONTINUOUS_TREATMENTS:
+                after_treatment_period_N = model_data.loc[model_data[after_treatment_col] > 0, time_col].nunique()
+            else:
+                after_treatment_period_N = model_data.loc[model_data[after_treatment_col] == 1, time_col].nunique()
             after_treatment_period_start = after_treatment_period_start.strftime(model_config["date_format"])
             after_treatment_period_end = after_treatment_period_end.strftime(model_config["date_format"])
             
@@ -1151,7 +1165,7 @@ class DiffModel:
     
     def prediction_intervals(
         self,
-        confint_alpha = 0.05
+        confint_alpha: float = 0.05
         ):
 
         """
@@ -1345,11 +1359,11 @@ class DiffModel:
         self,
         treatment: str = None,
         TG_col: str = None,
-        x_label = "Time",
-        y_label = "Analysis units",
+        x_label: str = "Time",
+        y_label: str = "Analysis units",
         y_lim = None,
-        plot_title = "Treatment time",
-        plot_symbol = "o",
+        plot_title: str = "Treatment time",
+        plot_symbol: str = "o",
         treatment_group_only = True
         ):
 
@@ -1422,7 +1436,12 @@ class DiffModel:
         modeldata_pivot.index = pd.to_datetime(modeldata_pivot.index)
 
         for i, col in enumerate(modeldata_pivot.columns):
-            time_points_treatment = modeldata_pivot.index[modeldata_pivot[col] == 1]
+            
+            if config.ACCEPT_CONTINUOUS_TREATMENTS:
+                time_points_treatment = modeldata_pivot.index[modeldata_pivot[col] > 0]
+            else:
+                time_points_treatment = modeldata_pivot.index[modeldata_pivot[col] == 1]
+            
             values = [i] * len(time_points_treatment)
             ax.plot(time_points_treatment, values, plot_symbol, label=col)
 
@@ -1466,7 +1485,8 @@ class DiffModel:
         plot_size: list = [12, 6],
         pre_post_ticks: list = ["Pre", "Post"],
         pre_post_barplot = False,
-        pre_post_bar_width = 0.5      
+        pre_post_bar_width = 0.5,
+        retransform_log_outcome: bool = False
         ):
 
         """
@@ -1510,6 +1530,9 @@ class DiffModel:
             Plot pre-post as barplot. Default is False.
         pre_post_bar_width : float, optional
             Bar width when pre_post_barplot is True. Default is 0.5.
+        retransform_log_outcome : bool, optional
+            If outcome was log-transformed, retransform to original scale for plotting. 
+            Default is False.
 
         Returns
         -------
@@ -1593,6 +1616,23 @@ class DiffModel:
     
         model_data = pd.concat ([model_data, model_predictions], axis = 1)
         
+        if retransform_log_outcome:
+
+            if model_config["log_outcome"]:
+
+                model_data[outcome_col] = np.exp(model_data[outcome_col])
+                model_data[outcome_col_predicted] = np.exp(model_data[outcome_col_predicted])
+
+                model_data[config.PREDICTIONS_SUMMARY_FRAME_COLS_LIST[2]] = np.exp(model_data[config.PREDICTIONS_SUMMARY_FRAME_COLS_LIST[2]])
+                model_data[config.PREDICTIONS_SUMMARY_FRAME_COLS_LIST[3]] = np.exp(model_data[config.PREDICTIONS_SUMMARY_FRAME_COLS_LIST[3]])
+                model_data[config.PREDICTIONS_SUMMARY_FRAME_COLS_LIST[4]] = np.exp(model_data[config.PREDICTIONS_SUMMARY_FRAME_COLS_LIST[4]])
+                model_data[config.PREDICTIONS_SUMMARY_FRAME_COLS_LIST[5]] = np.exp(model_data[config.PREDICTIONS_SUMMARY_FRAME_COLS_LIST[5]])
+                
+                print("NOTE: Outcome variable was log-transformed, Re-transformation was applied.")
+            
+            else:
+                print("NOTE: Parameter 'retransform_log_outcome' was set to True, but outcome variable was not log-transformed. No re-transformation applied.")
+                        
         model_data_TG = model_data[model_data[TG_col] == 1]
         model_data_CG = model_data[model_data[TG_col] == 0]
     
@@ -2230,16 +2270,20 @@ def did_analysis(
     treatment_diagnostics = treatment_diagnostics_results[0]
     staggered_adoption = treatment_diagnostics_results[1]
     
-    if no_treatments > 1:
-        FE_unit = True
+    if no_treatments > 1:        
+        
         intercept = False
-        TG_col = []
-        print("NOTE: Quasi-experiment includes more than one treatment. Unit fixed effects are used instead of control group baseline and treatment group deviation.")
-           
+        TG_col = []        
+        
+        if not FE_unit:
+            FE_unit = True
+            print("NOTE: Quasi-experiment includes more than one treatment. Unit fixed effects are used instead of control group baseline and treatment group deviation.")
+                       
     if ITE:
         
-        FE_unit = True
-        print("NOTE: Model includes individual treatment effects. Unit fixed effects are included.")
+        if not FE_unit:
+            FE_unit = True
+            print("NOTE: Model includes individual treatment effects. Unit fixed effects are included.")
         
         if GTE:
             GTE = False
@@ -2247,15 +2291,15 @@ def did_analysis(
     
     if ITT:
         
-        FE_unit = True
-        
         TT_col = []
         
-        print("NOTE: Model includes individual time trends. Unit fixed effects are included. Treatment time variable is dropped.")
+        if not FE_unit:
+            FE_unit = True
+            print("NOTE: Model includes individual time trends. Unit fixed effects are included. Treatment time variable is dropped.")
         
         if FE_time:
             FE_time = False
-            print("NOTE: Time fixed effects are dropped.")
+            print("NOTE: Model includes individual time trends. Time fixed effects are dropped.")
         
         if GTT:
             GTT = False
@@ -2263,10 +2307,11 @@ def did_analysis(
             
     if staggered_adoption:
         
+        if not FE_unit or not FE_time:
+            print("NOTE: Quasi-experiment includes one or more staggered treatments. Two-way fixed effects model is used.")
+            
         FE_unit = True
-        FE_time = True        
-        
-        print("NOTE: Quasi-experiment includes one or more staggered treatments. Two-way fixed effects model is used.")
+        FE_time = True      
     
     if GTT or GTE:        
         FE_group = True
@@ -2277,10 +2322,13 @@ def did_analysis(
     if FE_time:
         TT_col = []
         
-    if FE_group:    
+    if FE_group:
+        
+        if intercept or len(TG_col) > 0:
+            print("NOTE: Quasi-experiment includes group fixed effects. Control group baseline and treatment group deviation are dropped.")   
+            
         TG_col = []        
-        intercept = False
-        print("NOTE: Quasi-experiment includes group fixed effects. Control group baseline and treatment group deviation are dropped.")   
+        intercept = False        
     
     if after_treatment_col is not None or (isinstance (after_treatment_col, list) and len(after_treatment_col) > 0):
 

{diffindiff-2.3.0 → diffindiff-2.3.2}/diffindiff/diddata.py

@@ -4,8 +4,8 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     2.2.0
-# Last update: 2026-03-01 20:04
+# Version:     2.2.2
+# Last update: 2026-03-06 21:27
 # Copyright (c) 2024-2026 Thomas Wieland
 #-----------------------------------------------------------------------
 
@@ -174,7 +174,7 @@ class DiffGroups:
     def add_segmentation(
         self,
         group_benefit: list,
-        verbose: bool = config.VERBOSE       
+        verbose: bool = False       
         ):
         
         """
@@ -236,7 +236,7 @@ def create_groups(
     treatment_group,
     control_group,
     treatment_name: str = None,
-    verbose: bool = config.VERBOSE
+    verbose: bool = False
     ):
 
     """
@@ -538,7 +538,7 @@ def create_treatment(
     treatment_name: str = None,
     pre_post: bool = False,
     after_treatment_period: bool = False,
-    verbose = config.VERBOSE
+    verbose = False
     ): 
        
     """
@@ -1181,7 +1181,7 @@ class DiffData:
             print(f"WARNING: Additional data frame includes duplicate column names: {', '.join(existing_variables)}")
         
         return self
-
+   
     def add_treatment(
         self,
         treatment_name: str,
@@ -1190,7 +1190,7 @@ class DiffData:
         control_group,
         after_treatment_period: bool = False,
         after_treatment_name = None,
-        verbose: bool = config.VERBOSE
+        verbose: bool = False
         ):
         
         """
@@ -1389,7 +1389,7 @@ class DiffData:
         treatment_name: str,
         after_treatment_period: bool = False,
         after_treatment_name: str = None,
-        verbose: bool = config.VERBOSE
+        verbose: bool = False
         ):
 
         """
@@ -1420,25 +1420,25 @@ class DiffData:
 
         Examples
         --------
-        >>> Sensoria_Daten=did.create_data(
-        ...     outcome_data=Tourismusdaten_Tabelle,
-        ...     unit_id_col="Gemeinde",
-        ...     time_col="Jahr_Monat",
-        ...     outcome_col="Gaesteuebernachtungen_insgesamt",
-        ...     treatment_group=Tourismusdaten_Tabelle.loc[Tourismusdaten_Tabelle["Gemeinde"] == "255023 Holzminden,Stadt"]["Gemeinde"],
-        ...     control_group=Tourismusdaten_Tabelle.loc[Tourismusdaten_Tabelle["Gemeinde"] != "255023 Holzminden,Stadt"]["Gemeinde"],
-        ...     study_period=["2023-01-01", "2025-10-01"],
-        ...     treatment_period=["2024-09-01", "2025-10-01"],
-        ...     freq="MS",    
-        ...     treatment_name="Sensoria_geoeffnet",
+        >>> curfew_data_prepost=create_data(
+        ...     outcome_data=curfew_DE,
+        ...     unit_id_col="county",
+        ...     time_col="infection_date",
+        ...     outcome_col="infections_cum_per100000",
+        ...     treatment_group=curfew_DE.loc[curfew_DE["Bundesland"].isin([9,10,14])]["county"],
+        ...     control_group=curfew_DE.loc[~curfew_DE["Bundesland"].isin([9,10,14])]["county"],
+        ...     study_period=["2020-03-01", "2020-05-15"],
+        ...     treatment_period=["2020-03-21", "2020-05-05"],
+        ...     freq="D",
+        ...     pre_post=True
         ...     )
-        >>> Sensoria_Daten.add_covariates(
-        ...     additional_df=Tourismusdaten_Tabelle,
-        ...     variables=["Schlafgelegenheitentage_angeboten", "Event_Tage"],
-        ...     unit_col = "Gemeinde",
-        ...     time_col="Jahr_Monat",
+        >>> curfew_data_prepost_withcov = curfew_data_prepost.add_covariates(
+        ...     additional_df=counties_DE, 
+        ...     unit_col="county",
+        ...     time_col=None, 
+        ...     variables=["comm_index", "TourPer1000"]
         ...     )
-        >>> Sensoria_Daten.define_treatment("Event_Tage")
+        >>> curfew_data_prepost_withcov.define_treatment("TourPer1000")
         """
 
         if not treatment_name:            
@@ -1613,7 +1613,7 @@ class DiffData:
         counterfactual_outcome_col: str,
         time_col: str,
         counterfactual_UID: str = "counterfac",
-        verbose: bool = config.VERBOSE
+        verbose: bool = False
         ):
 
         """
@@ -1798,7 +1798,8 @@ class DiffData:
 
     def analysis(
         self, 
-        log_outcome: bool = False, 
+        log_outcome: bool = False,
+        log_outcome_add: float = 0.01,
         FE_unit: bool = False, 
         FE_time: bool = False,
         cluster_SE_by: str = None,
@@ -1814,7 +1815,7 @@ class DiffData:
         bonferroni: bool = False,
         drop_missing: bool = True,
         missing_replace_by_zero: bool = False,
-        verbose: bool = config.VERBOSE
+        verbose: bool = False
         ):
 
         """
@@ -1825,6 +1826,8 @@ class DiffData:
         ----------
         log_outcome : bool, optional
             If True, log-transform the outcome variable before estimation.
+        log_outcome_add : float, optional
+            Add a constant to outcome variable before log-transform.
         FE_unit : bool, optional
             Include unit fixed effects.
         FE_time : bool, optional
@@ -1914,7 +1917,7 @@ class DiffData:
                 BG_col = config.BG_COL,
                 outcome_col = outcome_col_original,
                 log_outcome = log_outcome,
-                log_outcome_add = 0.01,
+                log_outcome_add = log_outcome_add,
                 FE_unit = FE_unit,
                 FE_time = FE_time,                
                 covariates = covariates,
@@ -1961,6 +1964,7 @@ class DiffData:
                 ATT_col = ATT_col,
                 pre_post = treatment_meta["pre_post"],
                 log_outcome = log_outcome,
+                log_outcome_add = log_outcome_add,
                 FE_unit = FE_unit,
                 FE_time = FE_time,
                 cluster_SE_by = cluster_SE_by,
@@ -1994,7 +1998,7 @@ def merge_data(
     drop_missing: bool = True,
     missing_replace_by_zero: bool = False,
     keep_columns: bool = False,
-    verbose: bool = config.VERBOSE
+    verbose: bool = False
     ):
 
     """
@@ -2186,7 +2190,7 @@ def create_data(
     after_treatment_period: bool = False,
     drop_missing: bool = True,
     missing_replace_by_zero: bool = False,
-    verbose: bool = config.VERBOSE
+    verbose: bool = False
     ):
 
     """
@@ -2225,7 +2229,7 @@ def create_data(
     missing_replace_by_zero : bool, optional
         If True, replace missing values by zero when requested. Default is False.
     verbose : bool, optional
-        If True, print progress messages. Default follows package config.
+        If True, print progress messages.
 
     Returns
     -------

{diffindiff-2.3.0 → diffindiff-2.3.2}/diffindiff/didtools.py

@@ -4,8 +4,8 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     2.2.0
-# Last update: 2026-02-28 21:47
+# Version:     2.2.1
+# Last update: 2026-03-03 17:34
 # Copyright (c) 2025-2026 Thomas Wieland
 #-----------------------------------------------------------------------
 
@@ -434,7 +434,7 @@ def is_missing(
     if len(missing_true_vars) > 0:
         print(f"WARNING: Data frame contains columns with missing values: {', '.join(missing_true_vars)}.")
 
-    if drop_missing and not missing_replace_by_zero:
+    if drop_missing and not missing_replace_by_zero and len(missing_true_vars) > 0:
         
         if verbose:
             print("Dropping rows with missing values", end = " ... ")
@@ -521,9 +521,16 @@ def is_simultaneous(
         treatment_group = data_isnotreatment[1]
         data_TG = data[data[unit_col].isin(treatment_group)]
 
-        data_TG_pivot = data_TG.pivot_table (index = time_col, columns = unit_col, values = treatment_col)
+        data_TG_pivot = data_TG.pivot_table(
+            index = time_col,
+            columns = unit_col, 
+            values = treatment_col
+            )
 
-        simultaneous = (data_TG_pivot.nunique(axis=1) == 1).all()
+        if config.ACCEPT_CONTINUOUS_TREATMENTS:
+            simultaneous = (data_TG_pivot.nunique(axis=1) > 0).all()
+        else:
+            simultaneous = (data_TG_pivot.nunique(axis=1) == 1).all()
 
     if verbose:
         print("OK")
@@ -825,34 +832,67 @@ def is_parallel(
     
     treatment_group = modeldata_isnotreatment[1]
 
-    if len(data[(data[unit_col].isin(treatment_group)) & (data[treatment_col] == 1)]) > 0:
+    if config.ACCEPT_CONTINUOUS_TREATMENTS:
         
-        first_day_of_treatment = min(data[(data[unit_col].isin(treatment_group)) & (data[treatment_col] == 1)][time_col])
+        if len(data[(data[unit_col].isin(treatment_group)) & (data[treatment_col] > 0)]) > 0:
         
-        data_test = data[data[time_col] < first_day_of_treatment].copy()
-        data_test[config.TG_COL] = 0
-        data_test.loc[data_test[unit_col].isin(treatment_group), config.TG_COL] = 1
+            first_day_of_treatment = min(data[(data[unit_col].isin(treatment_group)) & (data[treatment_col] > 0)][time_col])
+            
+            data_test = data[data[time_col] < first_day_of_treatment].copy()
+            data_test[config.TG_COL] = 0
+            data_test.loc[data_test[unit_col].isin(treatment_group), config.TG_COL] = 1
+            
+            if config.TIME_COUNTER_COL not in data_test.columns:
+                data_test = date_counter(
+                    df = data_test,
+                    date_col = time_col, 
+                    new_col = config.TIME_COUNTER_COL,
+                    verbose = False
+                    )
+            data_test[f"{config.TG_COL}_x_{config.TIME_COL}"] = data_test[config.TG_COL]*data_test[config.TIME_COUNTER_COL]        
+
+            test_ols_model = ols(f'{outcome_col} ~ {config.TG_COL} + {config.TIME_COUNTER_COL} + {config.TG_COL}_x_{config.TIME_COL}', data = data_test).fit()
+            coef_TG_x_t_p = test_ols_model.pvalues[f"{config.TG_COL}_x_{config.TIME_COL}"]
+
+            if coef_TG_x_t_p < alpha:
+                parallel = False
+            else:
+                parallel = True
         
-        if config.TIME_COUNTER_COL not in data_test.columns:
-            data_test = date_counter(
-                df = data_test,
-                date_col = time_col, 
-                new_col = config.TIME_COUNTER_COL,
-                verbose = False
-                )
-        data_test[f"{config.TG_COL}_x_{config.TIME_COL}"] = data_test[config.TG_COL]*data_test[config.TIME_COUNTER_COL]        
-
-        test_ols_model = ols(f'{outcome_col} ~ {config.TG_COL} + {config.TIME_COUNTER_COL} + {config.TG_COL}_x_{config.TIME_COL}', data = data_test).fit()
-        coef_TG_x_t_p = test_ols_model.pvalues[f"{config.TG_COL}_x_{config.TIME_COL}"]
-
-        if coef_TG_x_t_p < alpha:
-            parallel = False
         else:
-            parallel = True
+            parallel = "not_tested"
+            test_ols_model = None
         
     else:
-        parallel = "not_tested"
-        test_ols_model = None
+
+        if len(data[(data[unit_col].isin(treatment_group)) & (data[treatment_col] == 1)]) > 0:
+            
+            first_day_of_treatment = min(data[(data[unit_col].isin(treatment_group)) & (data[treatment_col] == 1)][time_col])
+            
+            data_test = data[data[time_col] < first_day_of_treatment].copy()
+            data_test[config.TG_COL] = 0
+            data_test.loc[data_test[unit_col].isin(treatment_group), config.TG_COL] = 1
+            
+            if config.TIME_COUNTER_COL not in data_test.columns:
+                data_test = date_counter(
+                    df = data_test,
+                    date_col = time_col, 
+                    new_col = config.TIME_COUNTER_COL,
+                    verbose = False
+                    )
+            data_test[f"{config.TG_COL}_x_{config.TIME_COL}"] = data_test[config.TG_COL]*data_test[config.TIME_COUNTER_COL]        
+
+            test_ols_model = ols(f'{outcome_col} ~ {config.TG_COL} + {config.TIME_COUNTER_COL} + {config.TG_COL}_x_{config.TIME_COL}', data = data_test).fit()
+            coef_TG_x_t_p = test_ols_model.pvalues[f"{config.TG_COL}_x_{config.TIME_COL}"]
+
+            if coef_TG_x_t_p < alpha:
+                parallel = False
+            else:
+                parallel = True
+            
+        else:
+            parallel = "not_tested"
+            test_ols_model = None
     
     if verbose:
         print("OK")

{diffindiff-2.3.0 → diffindiff-2.3.2}/diffindiff.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: diffindiff
-Version: 2.3.0
+Version: 2.3.2
 Summary: diffindiff: Python library for convenient Difference-in-Differences analyses
 Author: Thomas Wieland
 Author-email: geowieland@googlemail.com
@@ -27,7 +27,7 @@ Thomas Wieland [ORCID](https://orcid.org/0000-0001-5168-9846) [EMail](mailto:geo
 
 If you use this software, please cite:
 
-Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.3.0) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
+Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.3.2) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
 
 
 ## Installation
@@ -52,10 +52,11 @@ pip install git+https://github.com/geowieland/diffindiff_official.git
   - Create ready-to-fit DiD data objects
   - Create predictive counterfactuals
 - **DiD analysis**: 
-  - Perfom standard DiD analysis with pre-post data
-  - Perfom DiD analysis with two-way fixed effects models
+  - Perform standard DiD analysis with pre-post data
+  - Perform DiD analysis with two-way fixed effects models
   - Simultaneous and/or staggered adoption are supported
   - Single or multiple treatments are supported
+  - Binary or continuous treatments are supported
   - Model extensions for DiD analysis:
     - Group- or individual-specific treatment effects
     - Group- or individual-specific time trends
@@ -66,10 +67,10 @@ pip install git+https://github.com/geowieland/diffindiff_official.git
   - Add own counterfactuals or create counterfactuals based on machine learning or OLS regression models
   - Bonferroni correction for treatment effects
   - Placebo test
-  - Test for control conditions
-  - Test for type of adoption
-  - Test whether the panel dataset is balanced
-  - Test for parallel trend assumption
+  - Test for control conditions (automatically within analysis or stand-alone)
+  - Test for type of adoption (automatically within analysis or stand-alone)
+  - Test whether the panel dataset is balanced (automatically within analysis or stand-alone)
+  - Test for parallel trend assumption (automatically within analysis or stand-alone)
 - **Visualization**:
   - Plot observed and expected time course of treatment and control group
   - Plot expected time course of treatment group and counterfactual
@@ -169,19 +170,10 @@ See the /tests directory for usage examples of most of the included functions.
 
 ## AI Usage Statement
 
-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.md` in the [public GitHub repository](https://github.com/geowieland/diffindiff_official).
+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 (v2.3.0)
+## What's new (v2.3.2)
 
-- General
-  - Full documentation (docstring in NumPy style) of all classes, methods, and functions
-  - Extended information in README
-- Extensions
-  - Timestamps for all methods creating or changing DiffGroups, DiffTreatment, DiffData and DiffModel objects
-  - Hyperparameters and other config information in didtools.model_wrapper() are now saved
-  - All summary methods return self
-- Bugfixes:
-  - Correct check of input lists in didanalysis_helper functions create_spillover() and extract_model_results() 
-  - Correct check of input lists in DiffData methods add_covariates() and analysis()
-  - Correct check of input lists in didanalysis functions did_analysis() and ddd_analysis()
+- Extensions:
+  - Re-transform log-transformed outcomes in DiffModel.plot() via parameter 'retransform_log_outcome'

{diffindiff-2.3.0 → diffindiff-2.3.2}/setup.py

@@ -7,7 +7,7 @@ def read_README():
     
 setup(
     name='diffindiff',
-    version='2.3.0',
+    version='2.3.2',
     description='diffindiff: Python library for convenient Difference-in-Differences analyses',
     packages=find_packages(include=["diffindiff", "diffindiff.tests"]),
     include_package_data=True,