contdid documentation

contdid is the checked Python release surface for continuous-treatment Difference-in-Differences estimators. The package keeps the public API traceable to the paper contracts, the reference R package, and the release verification artifacts in this repository.

Contents

• User Guide
  • Installation
  • Core Objects
  • Display Tables and Plots
  • Real-World Tutorial Provenance
  • Supported Public Routes
  • CCK Boundary
  • Data Rules
• Examples
  • Synthetic Dose Example
  • Worked Examples
  • Medicare Scaffold Walkthrough
  • Synthetic Dose Example
  • Release Examples
  • Real-world Medicare scaffold walkthrough
• API Reference
  • Public Package
  • Data Containers
  • Specifications
  • Results
  • Estimation
  • Event Study
  • Inference
  • Simulation
  • Empirical Scaffold

Status

The current Python package is an alpha release. The v1 surface includes checked input validation, synthetic data generation, parametric dose estimators, event-study routes, inference payload helpers, empirical scaffold metadata, and release-facing examples. Unsupported identification surfaces hard-fail rather than silently changing estimands.