causalis.scenarios.cuped.dgp.generate_cuped_tweedie_26generate_cuped_tweedie_26
Gold standard Tweedie-like DGP with mixed marginals and structured HTE.
This DGP generates a dataset representing a typical e-commerce scenario with many zeros and a heavy right tail (e.g., revenue). It features pre-treatment covariates (‘y_pre’, ‘y_pre_2’) that are calibrated to have a specific correlation with the post-treatment outcome.
Notes
The outcome $Y$ follows a Tweedie-like distribution where the mean is a function of treatment $D$ and latent factors $A$ and $X$:
The pre-period covariate $X_{pre}$ (e.g., y_pre) is generated to satisfy:
where $\rho$ is the pre_target_corr. This allows for realistic testing of
variance reduction techniques.
Examples
Generate data with two pre-period covariates
Parameters
- nint, default=10000
Number of samples to generate.
- seedint, default=42
Random seed.
- add_prebool, default=True
Whether to add pre-period covariates.
- pre_namestr, default=”y_pre”
Name of the first pre-period covariate column.
- pre_name_2str, optional
Name of the second pre-period covariate column. Defaults to
f"{pre_name}_2".- pre_target_corrfloat, default=0.82
Target correlation between the first pre covariate and post-outcome y in control group.
- pre_target_corr_2float, optional
Target correlation for the second pre covariate. Defaults to a moderate value based on
pre_target_corrto reduce collinearity.- pre_specPreCorrSpec, optional
Detailed specification for pre-period calibration (transform, method, etc.).
- include_oraclebool, default=False
Whether to include oracle ground-truth columns like ‘cate’, ‘propensity’, etc.
- return_causal_databool, default=True
Whether to return a CausalData object.
- theta_logfloat, default=0.38
The log-uplift theta parameter for the treatment effect.
Returns
pd.DataFrame or CausalData
Canonical target
causalis.scenarios.cuped.dgp.generate_cuped_tweedie_26
Sections