State society-wide reports run on the sparse national dataset

Aberta
  • Algo errado?
#3,864 0 comentários 0 reações 0 responsáveis Ver no GitHub

Mantenedores costumam responder em até 2 dias

Ninguém assumiu esta issue ainda.

  • #3868 de @anth-volk — fechado sem integrar

Avaliação

Dificuldade
4/5
Tempo estimado
3-5 dias
Facilidade para iniciantes
35/100
Tipo de issue
Bug
Clareza
Razoavelmente clara
Status de atividade
Ativa
Stack de tecnologia
python, typescript
Domínio
backend-api-design, data

Direção de pesquisa

Start with app/src/api/societyWideCalculation.ts and policyengine_api/services/economy_service.py, then read the tracked changes in PolicyEngine/policyengine.py#551 and PolicyEngine/policyengine-sim-api#717. Verify the API v1 bundle includes both upstream changes and that a region=state/pa economy result reports the local-area data version and dataset rather than the national build.

Escrita pelo modelo de indexação a partir do texto da issue.

Descrição

Problem

State society-wide reports in app-v2 run on the sparse national dataset.

  1. app-v2 calls /us/economy/{reform}/over/{baseline}?region=state/xx&time_period=YYYY and passes no dataset (app/src/api/societyWideCalculation.ts).
  2. EconomyService._build_economic_impact_setup_options rejects any non-default dataset (policyengine_api/services/economy_service.py lines 844–846) and submits dataset="default" (line 860).
  3. The simulation gateway resolves state/xx to the national file (PolicyEngine/policyengine-sim-api#717).

The deployed bundle is policyengine[models]==5.2.0 (pyproject.toml line 46), which uses policyengine-us 1.764.6 and data populace-us-2024-buildp-sparse-rmloss100-cae8640-20260728T011454Z. That matches what the app's report page shows ("Model version: 1.764.6 • Data version: populace-us-2024-buildp-sparse-rmloss100-cae8640-20260728T011454Z"). PolicyEngine.py 6.2.1 (PR #3855) moves to policyengine-us 2.2.1 and data populace-us-2024-spm-20260915. However, 6.2.1 also maps state regions to the national file, so state reports keep using national data after that bump.

Effect on a state report: Pennsylvania

With the deployed pair, the Pennsylvania slice has 1,999 household records and an effective sample of 314 (Kish ESS, 2026 baseline). Poor SPM units with children under 6 have an ESS of 5.2. A Pennsylvania refundable credit for children under 6 shows close to zero poverty impact at every amount up to $2,200 per child. The reason is that poor units with children under 6 whose poverty gap is at most $1,000 per child carry 0 weighted units, and at $2,200 per child they carry 2 (details in PolicyEngine/policyengine.py#551).

To track

  • PolicyEngine/policyengine.py#551 resolves state and district regions to the local-area dataset.
  • PolicyEngine/policyengine-sim-api#717 loads it for state and district runs.
  • API v1 installs a bundle that contains both changes.
  • A region=state/pa economy result reports the local-area build in policyengine_bundle.data_version / dataset, not the national build.

🤖 Generated with Claude Code

Linguagem predominante
Python
Estrelas
18
Forks
33
Merge médio
2d 9h
PRs com merge (30d)
77

Preparar o ambiente

Primeiros passos

  1. Leia a issue inteira e depois o guia de contribuição do projeto.
  2. Comente na issue dizendo que vai assumir — evita que duas pessoas façam o mesmo trabalho.
  3. Faça um fork do repositório e trabalhe em uma branch.
  4. Abra um pull request que referencie o número da issue.

Mais de PolicyEngine/policyengine-api

Todas as issues de PolicyEngine/policyengine-api

Issues semelhantes

Mais issues de Python

Receba novas issues na sua caixa de entrada

Um resumo curto de issues do GitHub para quem está começando.