Skip to content

Add 2025 ACA and Medicaid target coverage for calibration#663

Open
MaxGhenis wants to merge 1 commit intomainfrom
codex/ecps-2025-target-coverage
Open

Add 2025 ACA and Medicaid target coverage for calibration#663
MaxGhenis wants to merge 1 commit intomainfrom
codex/ecps-2025-target-coverage

Conversation

@MaxGhenis
Copy link
Copy Markdown
Contributor

Summary

  • add checked-in 2025 ACA and Medicaid target artifacts for the legacy calibration path
  • make legacy ACA/Medicaid target loading year-aware instead of hardcoding 2024 files
  • add 2022-to-2025 ACA state uprating multipliers so the unified matrix builder can use 2025 state factors
  • add regression tests for year selection and 2025 ACA factor loading

Why

This is PR 1 of the eCPS 2025 base-year migration tracked in #662. It does not switch the production dataset to 2025 yet. It makes the target layer ready for that move by ensuring a 2025 build can pull 2025 ACA and Medicaid artifacts instead of silently falling back to 2024 files.

Notes

  • For ACA, the 2025 state file uses CMS February 2025 APTC enrollment and average APTC by state. The legacy national 2025 ACA spending target is annualized from that monthly snapshot so the state and national ACA targets remain internally consistent.
  • For Medicaid, the 2025 state file uses the latest final-report month available in the March 2026 Medicaid.gov release (November 2025). The 2025 national Medicaid spending target uses the CMS NHE 2025 growth projection applied to 2024 Medicaid spending.

Testing

  • uv run ruff format --check policyengine_us_data/utils/loss.py policyengine_us_data/calibration/unified_matrix_builder.py policyengine_us_data/tests/test_calibration/test_loss_targets.py policyengine_us_data/tests/test_calibration/test_unified_matrix_builder.py
  • uv run pytest policyengine_us_data/tests/test_calibration/test_loss_targets.py policyengine_us_data/tests/test_calibration/test_unified_matrix_builder.py -q

Part of #662.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant