← Back to SE country layer · Country index

SE population QA cycle 2 distribution research: household/family/assignment/hidden priors

done synth-distributions-researcher

Task metadata

idt_f9075c12
titleSE population QA cycle 2 distribution research: household/family/assignment/hidden priors
assigneesynth-distributions-researcher
statusdone
tenantsynthestat
priority115
workspace_kinddir
workspace_path/home/synthestat
created_bysynth-manager
created_at2026-05-19 19:59:57 CEST
started_at2026-05-19 20:00:22 CEST
completed_at2026-05-19 20:06:13 CEST

Latest summary

Completed Sweden cycle-2 distribution prior research and wrote the requested SE population_distribution_priors handoff plus mirrored distribution findings/extraction/latest files. Key result: the claimed local sweden_household_composition bundle is absent and must be rebuilt/mirrored; SCB household/family tables are strong rebuild candidates, while family graph, assignments, shared custody/same-sex cohabitation, and hidden overlays must remain SOFT/GUIDE/partial with explicit uncertainty.

Body

Country: SE (Sweden)
Project root / allowed write root: /home/synthestat
Parent manager task: t_0c611b3b
Inputs to read first:
- /home/synthestat/workspace/manager_handoffs/SE_other_synthesis_ingest.md
- /home/synthestat/workspace/manager_handoffs/modeller/2026-05-19_1803_missing_requirements.md
- reviewer handoff on t_33ff07f7 (verdict NEEDS_MORE_SOURCES)
- docs/specs/research_knowledge_base.md
- docs/contracts/population_review_bundle.md

Mission:
Resolve Sweden distribution/prior evidence needed before the next SE modeler pass. Focus on household/family composition, relationship graph realism, origin/background assignment, work/school/second-home assignment priors, and hidden/weakly measured overlay priors.

Priority lanes:
1. Reconstruct or locate the Sweden household-composition prior bundle described in other_synthesis: data/distributions/sweden_household_composition/ with metadata/source_manifest.json, validation_report.md, table_usability.csv, priors/index.csv, direct/soft/historical priors, consolidated prior parquet/CSV/schema.
2. Validate concrete source candidates: HushallT05, HushallT07, HushallT28, LE0102T16N23, LE0102T17, LE0102T21N23, LE0102T23, GiftSkilsmassaTyp, CivilstandTypPar, MedelAlder1, Human Fertility Database, Kolk 2015, Kolk & Andersson 2019/2020, Mollborn/Kolk/Evertsson 2025, Hakovirta et al. 2023, Bergström et al. 2019, Helgertz et al. 2024, Cantalini/Ohlsson-Wijk/Andersson 2024.
3. Distinguish measured vs constrained vs modelled vs unknown for family_role, partner/parent links, sibling structures, same-sex cohabitation, shared custody/dual residence, first-birth age, parent-child/couple age gaps, income-standard bands.
4. Identify assignment priors for school, workplace, second homes only as distributions/priors; do not treat GUIDE mobility/Telia/POI signals as population-count measurements.
5. Hidden-population priors: determine if any defensible distributions exist for homelessness/refugees/Ukrainians/Syrians/undocumented/institutional/seasonal/student-housing; otherwise recommend explicit unavailable/partial overlays with uncertainty treatment.

Deliverable:
Write a concise distribution-research handoff under /home/synthestat/workspace/manager_handoffs/researcher/ named with timestamp and SE_population_distribution_priors. Include source IDs/URLs/citations, reference periods, geography levels, candidate constraint type (SOFT/GUIDE/INFORMATIONAL unless direct measured), uncertainty bounds or recommended interval strategy, and downloader-ready artifact requests.

Definition of done:
- Clear recommendation on whether the local household-prior bundle can be reused, must be rebuilt, or is absent.
- Concrete prior tables/papers are triaged into download now / citation only / licensed/unavailable / human decision.
- Explicit warnings for any prior that must not be represented as measured DeSO truth.

Parents

[
  "t_20de640c",
  "t_33ff07f7"
]

Children

[
  "t_1bbf9f63"
]

Runs

IDProfileStatusOutcomeStartedEndedSummary/error
79synth-distributions-researcherdonecompleted2026-05-19 20:00:22 CEST2026-05-19 20:06:13 CESTCompleted Sweden cycle-2 distribution prior research and wrote the requested SE population_distribution_priors handoff plus mirrored distribution findings/extraction/latest files. Key result: the claimed local sweden_household_composition bundle is absent and must be rebuilt/mirrored; SCB household/family tables are strong rebuild candidates, while family graph, assignments, shared custody/same-sex cohabitation, and hidden overlays must remain SOFT/GUIDE/partial with explicit uncertainty.

Events

TimeKindPayload
2026-05-19 19:59:57 CESTcreated{ "assignee": "synth-distributions-researcher", "status": "ready", "parents": [ "t_33ff07f7", "t_20de640c" ], "tenant": "synthestat", "skills": null }
2026-05-19 20:00:22 CESTclaimed{ "lock": "vmi3188806:1590352", "expires": 1779214522, "run_id": 79 }
2026-05-19 20:00:22 CESTspawned{ "pid": 1651416 }
2026-05-19 20:06:13 CESTcompleted{ "result_len": 0, "summary": "Completed Sweden cycle-2 distribution prior research and wrote the requested SE population_distribution_priors handoff plus mirrored distribution findings/extraction/latest files. Key result: the claimed local sweden_household_composition bundle is absent and must be rebuilt/mirrored; SCB household/family tables are strong rebuild candidates, while family graph, assignments, shared custody/same-se" }

Comments

No comments yet.