Controllable Generative Sandbox for Causal Inference
This work addresses the need for realistic and controllable synthetic data in causal inference, particularly for method validation and study design in domains like healthcare, though it is incremental as it builds on existing generative models with added causal controls.
The paper tackles the problem of generating synthetic data for causal inference that balances distributional realism with causal controllability, introducing CausalMix, a variational generative framework that achieves state-of-the-art distributional metrics on mixed-type tables while enabling independent manipulation of overlap, confounding strength, and treatment effect heterogeneity.
Method validation and study design in causal inference rely on synthetic data with known counterfactuals. Existing simulators trade off distributional realism, the ability to capture mixed-type and multimodal tabular data, against causal controllability, including explicit control over overlap, unmeasured confounding, and treatment effect heterogeneity. We introduce CausalMix, a variational generative framework that closes this gap by coupling a mixture of Gaussian latent priors with data-type-specific decoders for continuous, binary, and categorical variables. The model incorporates explicit causal controls: an overlap regularizer shaping propensity-score distributions, alongside direct parameterizations of confounding strength and effect heterogeneity. This unified objective preserves fidelity to the observed data while enabling factorial manipulation of causal mechanisms, allowing overlap, confounding strength, and treatment effect heterogeneity to be varied independently at design time. Across benchmarks, CausalMix achieves state-of-the-art distributional metrics on mixed-type tables while providing stable, fine-grained causal control. We demonstrate practical utility in a comparative safety study of metastatic castration-resistant prostate cancer treatments, using CausalMix to compare estimators under calibrated data-generating processes, tune hyperparameters, and conduct simulation-based power analyses under targeted treatment effect heterogeneity scenarios.