API Reference

This section provides complete API documentation for all diff-diff modules.

Estimators

Core estimator classes for DiD analysis:

`diff_diff.DifferenceInDifferences`	Difference-in-Differences estimator with sklearn-like interface.
`diff_diff.TwoWayFixedEffects`	Two-Way Fixed Effects (TWFE) estimator for panel DiD.
`diff_diff.MultiPeriodDiD`	Multi-Period Difference-in-Differences estimator.
`diff_diff.SyntheticDiD`	Synthetic Difference-in-Differences (SDID) estimator.
`diff_diff.CallawaySantAnna`	Callaway-Sant'Anna (2021) estimator for staggered Difference-in-Differences.
`diff_diff.SunAbraham`	Sun-Abraham (2021) interaction-weighted estimator for staggered DiD.
`diff_diff.ImputationDiD`	Borusyak-Jaravel-Spiess (2024) imputation DiD estimator.
`diff_diff.StackedDiD`	Stacked Difference-in-Differences estimator.
`diff_diff.TripleDifference`	Triple Difference (DDD) estimator.
`diff_diff.TROP`	Triply Robust Panel (TROP) estimator.

Result containers returned by estimators:

`diff_diff.DiDResults`	Results from a Difference-in-Differences estimation.
`diff_diff.MultiPeriodDiDResults`	Results from a Multi-Period Difference-in-Differences estimation.
`diff_diff.SyntheticDiDResults`	Results from a Synthetic Difference-in-Differences estimation.
`diff_diff.PeriodEffect`	Treatment effect for a single time period.
`diff_diff.CallawaySantAnnaResults`	Results from Callaway-Sant'Anna (2021) staggered DiD estimation.
`diff_diff.CSBootstrapResults`	Results from Callaway-Sant'Anna multiplier bootstrap inference.
`diff_diff.GroupTimeEffect`	Treatment effect for a specific group-time combination.
`diff_diff.SunAbrahamResults`	Results from Sun-Abraham (2021) interaction-weighted estimation.
`diff_diff.SABootstrapResults`	Results from Sun-Abraham bootstrap inference.
`diff_diff.ImputationDiDResults`	Results from Borusyak-Jaravel-Spiess (2024) imputation DiD estimation.
`diff_diff.ImputationBootstrapResults`	Results from ImputationDiD bootstrap inference.
`diff_diff.TripleDifferenceResults`	Results from Triple Difference (DDD) estimation.
`diff_diff.StackedDiDResults`	Results from Stacked DiD estimation (Wing, Freedman & Hollingsworth 2024).
`diff_diff.trop.TROPResults`	Results from a Triply Robust Panel (TROP) estimation.

Plotting functions for results:

`diff_diff.plot_event_study`	Create an event study plot showing treatment effects over time.
`diff_diff.plot_group_effects`	Plot treatment effects by treatment cohort (group).
`diff_diff.plot_sensitivity`	Plot sensitivity analysis results from Honest DiD.
`diff_diff.plot_honest_event_study`	Create event study plot with Honest DiD confidence intervals.
`diff_diff.plot_bacon`	Visualize Goodman-Bacon decomposition results.
`diff_diff.plot_power_curve`	Create a power curve visualization.
`diff_diff.plot_pretrends_power`	Plot pre-trends test power curve.

Placebo tests and model diagnostics:

`diff_diff.run_placebo_test`	Run a placebo test to validate DiD assumptions.
`diff_diff.placebo_timing_test`	Test for pre-treatment effects by moving treatment timing earlier.
`diff_diff.placebo_group_test`	Test for differential trends among never-treated units.
`diff_diff.permutation_test`	Compute permutation-based p-value for DiD estimate.
`diff_diff.leave_one_out_test`	Assess sensitivity by dropping each treated unit in turn.
`diff_diff.run_all_placebo_tests`	Run a comprehensive suite of placebo tests.
`diff_diff.PlaceboTestResults`	Results from a placebo test for DiD assumption validation.

Honest DiD for robust inference:

`diff_diff.HonestDiD`	Honest DiD sensitivity analysis (Rambachan & Roth 2023).
`diff_diff.HonestDiDResults`	Results from Honest DiD sensitivity analysis.
`diff_diff.SensitivityResults`	Results from sensitivity analysis over a grid of M values.
`diff_diff.DeltaSD`	Smoothness restriction on trend violations (Delta^{SD}).
`diff_diff.DeltaRM`	Relative magnitudes restriction on trend violations (Delta^{RM}).
`diff_diff.DeltaSDRM`	Combined smoothness and relative magnitudes restriction.
`diff_diff.compute_honest_did`	Convenience function for computing Honest DiD bounds.
`diff_diff.sensitivity_plot`	Create a sensitivity analysis plot.

Testing the parallel trends assumption:

`diff_diff.check_parallel_trends`	Perform a simple check for parallel trends assumption.
`diff_diff.check_parallel_trends_robust`	Perform robust parallel trends testing using distributional comparisons.
`diff_diff.equivalence_test_trends`	Perform equivalence testing (TOST) for parallel trends.

Wild cluster bootstrap for valid inference:

`diff_diff.wild_bootstrap_se`	Compute wild cluster bootstrap standard errors and p-values.
`diff_diff.WildBootstrapResults`	Results from wild cluster bootstrap inference.

Power analysis for study design:

`diff_diff.PowerAnalysis`	Power analysis for difference-in-differences designs.
`diff_diff.PowerResults`	Results from analytical power analysis.
`diff_diff.SimulationPowerResults`	Results from simulation-based power analysis.
`diff_diff.compute_power`	Convenience function to compute power for given effect and sample.
`diff_diff.compute_mde`	Convenience function to compute minimum detectable effect.
`diff_diff.compute_sample_size`	Convenience function to compute required sample size.
`diff_diff.simulate_power`	Estimate power using Monte Carlo simulation.

Power analysis for pre-trends tests (Roth 2022):

`diff_diff.PreTrendsPower`	Pre-trends power analysis (Roth 2022).
`diff_diff.PreTrendsPowerResults`	Results from pre-trends power analysis.
`diff_diff.PreTrendsPowerCurve`	Power curve across violation magnitudes.
`diff_diff.compute_pretrends_power`	Convenience function for pre-trends power analysis.
`diff_diff.compute_mdv`	Compute minimum detectable violation.

Utilities for preparing DiD data:

`diff_diff.generate_did_data`	Generate synthetic data for DiD analysis with known treatment effect.
`diff_diff.make_treatment_indicator`	Create a binary treatment indicator column from various input types.
`diff_diff.make_post_indicator`	Create a binary post-treatment indicator column.
`diff_diff.wide_to_long`	Convert wide-format panel data to long format for DiD analysis.
`diff_diff.balance_panel`	Balance a panel dataset to ensure all units have all time periods.
`diff_diff.validate_did_data`	Validate that data is properly formatted for DiD analysis.
`diff_diff.summarize_did_data`	Generate summary statistics by treatment group and time period.
`diff_diff.create_event_time`	Create an event-time column relative to treatment timing.
`diff_diff.aggregate_to_cohorts`	Aggregate unit-level data to treatment cohort means.
`diff_diff.rank_control_units`	Rank potential control units by their suitability for DiD analysis.

Detailed documentation by module: