diffindiff 1.2.0__tar.gz → 1.2.2__tar.gz

{diffindiff-1.2.0 → diffindiff-1.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: diffindiff
-Version: 1.2.0
+Version: 1.2.2
 Summary: diffindiff: Python library for convenient Difference-in-Differences Analyses
 Author: Thomas Wieland
 Author-email: geowieland@googlemail.com
@@ -10,7 +10,7 @@ Requires-Dist: pandas
 Requires-Dist: statsmodels
 Requires-Dist: matplotlib
 Requires-Dist: datetime
-Requires-Dist: sklearn
+Requires-Dist: scikit-learn
 Requires-Dist: xgboost
 Requires-Dist: lightgbm
 Dynamic: author
@@ -24,6 +24,8 @@ Dynamic: summary
 
 This Python library is designed for performing Difference-in-Differences (DiD) analyses in a convenient way. It allows users to construct datasets, define treatment and control groups, and set treatment periods. DiD model analyses may be conducted with both datasets created by built-in functions and ready-to-use external datasets. Both simultaneous and staggered adoption are supported. The library allows for various extensions, such as two-way fixed effects models, group- or individual-specific effects, and post-treatment periods. Additionally, it includes functions for visualizing results, such as plotting DiD coefficients with confidence intervals and illustrating the temporal evolution of staggered treatments.
 
+The diffindiff package is currently being extensively revised (v2.0.0 and following).
+
 
 ## Author
 
@@ -46,6 +48,7 @@ Thomas Wieland [ORCID](https://orcid.org/0000-0001-5168-9846) [EMail](mailto:geo
     - Including covariates
     - After-treatment period
     - Triple Difference (DDD)
+    - Multiple treatments
     - Own counterfactuals
 - **Visualization**: e.g.
   - Plot observed and expected time course of treatment and control group

{diffindiff-1.2.0 → diffindiff-1.2.2}/README.md

@@ -2,6 +2,8 @@
 
 This Python library is designed for performing Difference-in-Differences (DiD) analyses in a convenient way. It allows users to construct datasets, define treatment and control groups, and set treatment periods. DiD model analyses may be conducted with both datasets created by built-in functions and ready-to-use external datasets. Both simultaneous and staggered adoption are supported. The library allows for various extensions, such as two-way fixed effects models, group- or individual-specific effects, and post-treatment periods. Additionally, it includes functions for visualizing results, such as plotting DiD coefficients with confidence intervals and illustrating the temporal evolution of staggered treatments.
 
+The diffindiff package is currently being extensively revised (v2.0.0 and following).
+
 
 ## Author
 
@@ -24,6 +26,7 @@ Thomas Wieland [ORCID](https://orcid.org/0000-0001-5168-9846) [EMail](mailto:geo
     - Including covariates
     - After-treatment period
     - Triple Difference (DDD)
+    - Multiple treatments
     - Own counterfactuals
 - **Visualization**: e.g.
   - Plot observed and expected time course of treatment and control group

{diffindiff-1.2.0 → diffindiff-1.2.2}/diffindiff/didanalysis.py

@@ -2,8 +2,8 @@
 # Name:        didanalysis (diffindiff)
 # Purpose:     Analysis functions for difference-in-differences analyses
 # Author:      Thomas Wieland (geowieland@googlemail.com)
-# Version:     1.2.0
-# Last update: 2025-03-25 19:30
+# Version:     1.2.2
+# Last update: 2025-04-03 06:55
 # Copyright (c) 2025 Thomas Wieland
 #-------------------------------------------------------------------------------
 
@@ -13,6 +13,7 @@ from statsmodels.formula.api import ols
 import numpy as np
 import matplotlib.pyplot as plt
 from matplotlib.dates import DateFormatter
+import warnings
 from diffindiff import didtools
 
 
@@ -111,6 +112,8 @@ class did_model:
         self,
         full = True
         ):
+
+        warnings.warn("This function will be fundamentally changed in a future version (v2.0.0 and subsequent) of the diffindiff package.", FutureWarning)
         
         model_results = self.data[0]
         model_config = self.data[1]
@@ -497,6 +500,8 @@ class did_model:
 
     def effects(self):
 
+        warnings.warn("This function will be replaced by the treatment_effects() function in a future version (v2.0.0 and subsequent versions) of the diffindiff package.", FutureWarning)
+
         model_results = self.data[0]
 
         effects_df = pd.DataFrame (columns = ["effect_name", "coef", "SE", "t", "p", "CI_lower", "CI_upper"])
@@ -575,15 +580,21 @@ class did_model:
 
     def fixef(self):
 
+        warnings.warn("This function is deprecated and will be removed in a future version (v2.0.0 and subsequent) of the diffindiff package.", FutureWarning)
+
         fixed_effects = self.data[4]
         return fixed_effects
 
     def indef(self):
 
+        warnings.warn("This function is deprecated and will be removed in a future version (v2.0.0 and subsequent) of the diffindiff package.", FutureWarning)
+
         individual_effects = self.data[5]
         return individual_effects
 
     def groupef(self):
+        
+        warnings.warn("This function is deprecated and will be removed in a future version (v2.0.0 and subsequent) of the diffindiff package.", FutureWarning)
 
         group_effects = self.data[6]
         return group_effects
@@ -981,7 +992,9 @@ class did_model:
         plot_size: list = [7, 6],      
         scale_plot: bool = True
         ):
-        
+
+        warnings.warn("This function is deprecated and will be removed in a future version (v2.0.0 and subsequent) of the diffindiff package.", FutureWarning)
+
         effects = self.effects()
 
         if sort_by == "coef":
@@ -1035,6 +1048,8 @@ class did_model:
             central_tendency = "mean"
             ):
 
+        warnings.warn("This function is deprecated and will be removed in a future version (v2.0.0 and subsequent) of the diffindiff package.", FutureWarning)
+
         model_config = self.data[1]
         if not model_config["GTE"]:
             raise ValueError ("Model does not include group treatment effects. Set GTE=True an define grouping variable using group_by.")
@@ -1101,6 +1116,8 @@ class did_model:
             show_central_tendency = False,
             central_tendency = "mean"
             ):
+        
+        warnings.warn("This function is deprecated and will be removed in a future version (v2.0.0 and subsequent) of the diffindiff package.", FutureWarning)
 
         model_config = self.data[1]
         if not model_config["ITE"]:

{diffindiff-1.2.0 → diffindiff-1.2.2}/diffindiff/diddata.py

@@ -2,8 +2,8 @@
 # Name:        diddata (diffindiff)
 # Purpose:     Creating data for Difference-in-Differences Analysis
 # Author:      Thomas Wieland (geowieland@googlemail.com)
-# Version:     1.2.0
-# Last update: 2025-03-25 19:28
+# Version:     1.2.2
+# Last update: 2025-04-03 06:59
 # Copyright (c) 2025 Thomas Wieland
 #-------------------------------------------------------------------------------
 
@@ -359,11 +359,6 @@ class did_data:
 
         return self
 
-    def add_treatment( # TODO ?? hier Funktion, die 1 weiteres Treament hinzufuegt
-        self
-        ):
-        pass
-
     def add_segmentation(
         self,
         group_benefit: list
@@ -642,22 +637,6 @@ def create_data(
 
     return did_data_all
 
-class counterfactual: # TODO ??
-    def __init__(
-        self, 
-        model_results 
-        
-        ):
-
-        self.data = [
-            model_results
-            ]
-
-    def get_modelresults (self):
-        return self.data[0]
-    
-
-
 def create_counterfactual(
     data,
     y: str,
@@ -708,7 +687,7 @@ def create_counterfactual(
     units = didtools.unique(units_tt[unit_col])  
     
     if not isnotreatment[0]:
-        print ("No no-treatment control group")    
+        print ("No no-treatment control group. Counterfactual will not cover full treatment time.")    
     
     data_TG = pd.DataFrame(columns = data.columns)
     for unit in units:

{diffindiff-1.2.0 → diffindiff-1.2.2}/diffindiff/didtools.py

@@ -2,8 +2,8 @@
 # Name:        didtools (diffindiff)
 # Purpose:     Creating data for Difference-in-Differences Analysis
 # Author:      Thomas Wieland (geowieland@googlemail.com)
-# Version:     1.2.0
-# Last update: 2025-03-25 19:27
+# Version:     1.2.2
+# Last update: 2025-04-03 06:54
 # Copyright (c) 2025 Thomas Wieland
 #-------------------------------------------------------------------------------
 
@@ -143,41 +143,47 @@ def is_parallel(
     pre_post = False,
     alpha = 0.05
     ):
-
-    if pre_post:
-        return None
-    
+ 
     modeldata_isnotreatment = is_notreatment(
         data = data,
         unit_col = unit_col,
         treatment_col = treatment_col
-        )    
-    if not modeldata_isnotreatment:
-        return None
+        )
+        
+    if pre_post or not modeldata_isnotreatment:
+        parallel = "not_tested"
+        test_ols_model = None
     
     treatment_group = modeldata_isnotreatment[1]
-    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["TG"] = 0
-    data_test.loc[data_test[unit_col].isin(treatment_group), "TG"] = 1
-    
-    if "date_counter" not in data_test.columns:
-        data_test = date_counter(
-            df = data_test,
-            date_col = time_col, 
-            new_col = "date_counter"
-            )
-    data_test["TG_x_t"] = data_test["TG"]*data_test["date_counter"]
-
-    test_ols_model = ols(f'{outcome_col} ~ TG + date_counter + TG_x_t', data = data_test).fit()
-    coef_TG_x_t_p = test_ols_model.pvalues["TG_x_t"]
-
-    if coef_TG_x_t_p < alpha:
-        parallel = False
+    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["TG"] = 0
+        data_test.loc[data_test[unit_col].isin(treatment_group), "TG"] = 1
+        
+        if "date_counter" not in data_test.columns:
+            data_test = date_counter(
+                df = data_test,
+                date_col = time_col, 
+                new_col = "date_counter"
+                )
+        data_test["TG_x_t"] = data_test["TG"]*data_test["date_counter"]
+
+        test_ols_model = ols(f'{outcome_col} ~ TG + date_counter + TG_x_t', data = data_test).fit()
+        coef_TG_x_t_p = test_ols_model.pvalues["TG_x_t"]
+
+        if coef_TG_x_t_p < alpha:
+            parallel = False
+        else:
+            parallel = True
+        
     else:
-        parallel = True
-
+        parallel = "not_tested"
+        test_ols_model = None
+        
     return [
         parallel, 
         test_ols_model

{diffindiff-1.2.0 → diffindiff-1.2.2}/diffindiff.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: diffindiff
-Version: 1.2.0
+Version: 1.2.2
 Summary: diffindiff: Python library for convenient Difference-in-Differences Analyses
 Author: Thomas Wieland
 Author-email: geowieland@googlemail.com
@@ -10,7 +10,7 @@ Requires-Dist: pandas
 Requires-Dist: statsmodels
 Requires-Dist: matplotlib
 Requires-Dist: datetime
-Requires-Dist: sklearn
+Requires-Dist: scikit-learn
 Requires-Dist: xgboost
 Requires-Dist: lightgbm
 Dynamic: author
@@ -24,6 +24,8 @@ Dynamic: summary
 
 This Python library is designed for performing Difference-in-Differences (DiD) analyses in a convenient way. It allows users to construct datasets, define treatment and control groups, and set treatment periods. DiD model analyses may be conducted with both datasets created by built-in functions and ready-to-use external datasets. Both simultaneous and staggered adoption are supported. The library allows for various extensions, such as two-way fixed effects models, group- or individual-specific effects, and post-treatment periods. Additionally, it includes functions for visualizing results, such as plotting DiD coefficients with confidence intervals and illustrating the temporal evolution of staggered treatments.
 
+The diffindiff package is currently being extensively revised (v2.0.0 and following).
+
 
 ## Author
 
@@ -46,6 +48,7 @@ Thomas Wieland [ORCID](https://orcid.org/0000-0001-5168-9846) [EMail](mailto:geo
     - Including covariates
     - After-treatment period
     - Triple Difference (DDD)
+    - Multiple treatments
     - Own counterfactuals
 - **Visualization**: e.g.
   - Plot observed and expected time course of treatment and control group

{diffindiff-1.2.0 → diffindiff-1.2.2}/diffindiff.egg-info/requires.txt

@@ -3,6 +3,6 @@ pandas
 statsmodels
 matplotlib
 datetime
-sklearn
+scikit-learn
 xgboost
 lightgbm

{diffindiff-1.2.0 → diffindiff-1.2.2}/setup.py

@@ -7,7 +7,7 @@ def read_README():
     
 setup(
     name='diffindiff',
-    version='1.2.0',
+    version='1.2.2',
     description='diffindiff: Python library for convenient Difference-in-Differences Analyses',
     packages=find_packages(include=["diffindiff", "diffindiff.tests"]),
     include_package_data=True,
@@ -25,7 +25,7 @@ setup(
         'statsmodels',
         'matplotlib',
         'datetime',
-        'sklearn',
+        'scikit-learn',
         'xgboost',
         'lightgbm'
     ],