diffindiff Release of version 2.3.7

What's new (v2.3.7)

• Extensions:

  • DiffModel.counterfactual() and DiffModel.plot_counterfactual(): Re-transform log-transformed outcome via parameter retransform_log_outcome
  • Difference between predicted and counterfactual-predicted in result df of DiffModel.counterfactual()
  • Extended parameters in didtools.model_wrapper() to be passed to sklearn.model_selection.train_test_split()
  • Extension of documentations

• Bugfixes:

  • diddata.create_counterfactual(): Fixed calculation error, additional check of input parameters
  • didtools.is_prepost() now also checks number of observational units
  • didtools.is_notreatment() now checks whether data is singular with respect to 0 and 1
  • didtools.fit_metrics() tries to convert observed and expected if not numeric before raising TypeError

diffindiff Release of version 2.3.6

What's new (v2.3.6)

• Bugfixes:
  • Bug in DiffModel.placebo() with respect to missing treatment period and time columns fixed
  • Collecting treatment column errors in DiffModel.placebo()
  • Corrections in error messages

diffindiff Release of version 2.3.5

What's new (v2.3.5)

• Bugfixes:
  • Test whether input data is panel data via didtools.is_panel() which is included in didanalysis_helper.data_diagnostics()
  • Fixed false test results given continuous treatments are accepted in didtools.is_simultaneous()
  • Fixed false test results given continuous treatments are accepted in didtools.is_prepost()
  • Argument 'pre_post' is passed to is_simultaneous() in didanalysis_helper.treatment_diagnostics()

diffindiff Release of version 2.3.4

What's new (v2.3.4)

• Bugfixes:

  • Fixed bug in DiffData instance creation in diddata.merge_data()

diffindiff Release of version 2.3.3

What's new (v2.3.3)

• Extensions:

  • Create treatment interactions via didanalysis.did_analysis() or diddata.DiffData.analysis()

diffindiff Release of version 2.3.2

What's new (v2.3.2)

• Extensions:
  • Re-transform log-transformed outcomes in DiffModel.plot() via parameter 'retransform_log_outcome'

diffindiff Release of version 2.3.1

What's new (v2.3.1)

• Bugfixes:

  • didtools.is_missing(): No "Removing NA" note anymore when no missing values occur
  • didanalyis.did_analysis(): No "xx fixed effects included" notes anymore when FE_unit and/or FE_time are already set to True by user
  • DiffModel methods treatment_statistics() and plot_timeline() now accept continuous treatments (value of treatment variable > 0)
  • didtools function is_parallel() now accepts continuous treatments (value of treatment variable > 0)
  • Added parameter log_outcome_add in method DiffData.analysis()
  • Corrected examples in docstrings
  • Corrected typos in README

diffindiff Release of version 2.3.0

What's new (v2.3.0)

• 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()

diffindiff Release of version 2.2.7

What's new (v2.2.7)

• Functions

  • diddata.DiffData.define_treatment() for constructing a new treatment from a column in the dataframe

• Bugfixes:

  • didtools.treatment_times() and didtools.is_multiple_treatment_period() now also identify continuous treatments correctly
  • Fixed problematic type conversion in didtools.fit_metrics()

diffindiff: Python library for convenient Difference-in-Differences analyses

What's new (v2.2.6)

• Bugfixes:
  • Check for correct dates in diddata.create_treatment()
  • Check for valid columns in diddata.merge_data()
  • Removed unnecessary old dependencies and imports
• Other:
  • Changed diddata.DiffGroups.add_segmentation() to return a message rather than raising an exception when the DiffGroups object already includes a benefit group