causalis.scenarios.did.dgp API Reference | Causalis

data

causalis.scenarios.did.dgp.PanelOutput

PanelOutput

Value: None

None

Canonical target

causalis.scenarios.did.dgp.PanelOutput

Link to this symbol

function

causalis.scenarios.did.dgp.generate_did_gamma_26

generate_did_gamma_26

Generate a staggered-adoption Gamma DID panel for CSA estimators.

The returned panel is long-format with absorbing treatment, multiple first-treatment cohorts, never-treated controls, baseline-compatible covariates, cluster labels, and optional oracle counterfactual/effect columns. It is intended to be consumed directly by :class:causalis.scenarios.did.model.CallawaySantAnnaDID.

Parameters

seedint, default 42: Random seed for reproducibility.
return_panel_databool, default True: If True, returns a :class:PanelDataDID contract; if False, returns a raw :class:pd.DataFrame.
include_oraclesbool, default True: Whether to include ground-truth columns (e.g., y_cf, tau_mean_true).
n_treated_unitsint, default 20: Number of units that will eventually receive treatment.
n_control_unitsint, default 60: Number of never-treated units.
n_pre_periodsint, optional: Number of periods before any treatment starts. Defaults to 24.
n_post_periodsint, optional: Number of periods after the first treatment cohort starts. Defaults to 8.
n_cohortsint, default 4: Number of distinct treatment-start times (cohorts) among treated units.
treatment_effect_ratefloat, default 0.08: Base treatment effect as a fraction of the counterfactual outcome.
treatment_effect_slopefloat, default 0.0: The rate at which the treatment effect grows or decays per period after start.
treatment_effect_heterogeneity_stdfloat, default 0.025: Standard deviation of unit-level treatment effect multipliers.
parallel_trend_violationfloat, default 0.0: Strength of a differential trend between treated and control units (0 = parallel).
calendar_startstr, default “2021-01”: The starting period string for the pandas index.
time_freqstr, default “M”: The pandas frequency alias (e.g., “M”, “W”, “D”).
treated_prefixstr, default “treated_market_”: Prefix for treated unit IDs.
control_prefixstr, default “control_market_”: Prefix for control unit IDs.
**advanced_paramsdict: Reserved for future parameters.

Returns

PanelDataDID or pd.DataFrame

The generated panel dataset.

Examples

Generate default panel data

Notes

The DGP simulates a complex business environment where the outcome $Y_{it}$ (e.g., revenue) follows a Gamma distribution:

Y_{it} \sim \text{Gamma}\left( \text{shape}=\kappa, \text{scale}=\frac{\mu_{it}}{\kappa} \right)

The mean $\mu_{it}$ is decomposed into a counterfactual $\mu_{it}(0)$ and a treatment effect $\tau_{it}$ :

\mu_{it}(1) = \mu_{it}(0) \cdot (1 + \tau_{it})

The counterfactual mean $\mu_{it}(0)$ is modeled as a product of exposure, conversion rate, and average order value (AOV), each subject to macro-economic cycles, seasonality, and unit-level trends:

\ln \mu_{it}(0) = \ln(\text{Exposure}_{it}) + \ln(\text{ConvRate}_{it}) + \ln(\text{AOV}_{it})

where each component follows an AR(1) process with latent confounding. The treatment effect $\tau_{it}$ for a unit $i$ treated at time $G_i$ is:

\tau_{it} = (\theta_{rate} + \theta_{slope} \cdot (t - G_i)) \cdot \text{Ramp}(t - G_i) \cdot \eta_i

where $\text{Ramp}(\cdot)$ is an exponential adoption curve and $\eta_i$ is unit-level heterogeneity. Parallel trend violations are introduced by adding a differential linear trend to treated units’ counterfactuals.

Canonical target

causalis.scenarios.did.dgp.generate_did_gamma_26

Sections

ParametersReturnsNotesExamples

Link to this symbol

data

causalis.scenarios.did.dgp.generate_staggered_did_gamma_26

generate_staggered_did_gamma_26

Value: None

None

Canonical target

causalis.scenarios.did.dgp.generate_staggered_did_gamma_26

Link to this symbol

data

causalis.scenarios.did.dgp.__all__

all

Value: ['generate_did_gamma_26', 'generate_staggered_did_gamma_26']

[‘generate_did_gamma_26’, ‘generate_staggered_did_gamma_26’]

Canonical target

causalis.scenarios.did.dgp.__all__

Link to this symbol