PolicyEngine / PolicyEngine/microcosm

SIPP+SCF blend + SSI-aware refit: land the asset magnitudes for the Build J re-certification

Open
#374 7 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
0
Forks
4
Avg merge
1d 3h
Merged PRs (30d)
94

Description

Follow-up to #373 (SCF wealth stage): reach the dense-native SSI magnitude before the Build J re-certification flip.

Two steps, per the #373 probe diagnosis:

  1. SIPP blend: the retired pipeline mixed SIPP and SCF financial-asset donors at FINANCIAL_ASSET_SOURCE_SCF_PROBABILITY = 0.5 (retired asset_imputation.py) — SIPP carries realistic low liquid assets; SCF-only overstates the SSI-relevant left tail. Port the SIPP donor half.
  2. SSI-aware refit: the +$9.66B probe ran as a post-hoc overlay on weights solved without assets (the #356-documented +$9-26B overlay band). The certified rebuild must re-solve weights with the asset columns present (and an SSI baseline target if adopted), which the #356 dense-native record shows lands $10k/$20k at ~+$1.60B.

Acceptance: the #369 reform-coverage smoke stays green AND the $10k/$20k delta lands in the dense-native reference class on the re-certified artifact.

Refs: #368, #373, #356, #369.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the retired asset_imputation.py approach described in the issue and read the #373 probe diagnosis, then compare the #356 dense-native record. Port the SIPP donor half and perform the SSI-aware certified rebuild; done means the #369 reform-coverage smoke remains green and the $10k/$20k delta matches the dense-native reference class.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
data
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
42/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.