Build a Staggered Policy Rollout Impact Study Workspace — Implementation planning and evidence-linked development

## User
An applied policy researcher evaluating a phased program.

## Build
Study a policy introduced at different times across locations. Make comparison groups, timing assumptions, pretrends, and uncertainty visible before presenting an effect estimate.

## Input Contract
- Key: region_id + period.
- Fields: outcome, treatment_start, covariates, population_weight, neighboring_exposure.
- Record anticipation and spillover assumptions. Never use already-treated regions as untreated without justification.

## Workflow
1. Define the estimand and eligible comparison cohorts.
2. Inspect event-time patterns and treatment timing.
3. Estimate cohort-aware effects with appropriate clustered uncertainty.
4. Export robustness results and an identification decision log.

## Interface
A rollout map, cohort comparison editor, event-study plot, and assumptions review panel.

## Deliverables
- A reproducible cohort-based impact workflow.
- A report with estimands, comparison rules, uncertainty, and falsification results.
- Runnable local application, setup commands, and a dependency lockfile.
- Versioned input data, data dictionary, acquisition instructions, and rights notes.
- Calculation traces, source references, assumption records, and exported review artifacts.
- A deterministic baseline and tests for normal, incomplete, contradictory, and failure-recovery cases.
- A recorded complete workflow with a reproducible manifest and explicit limitations.

## Acceptance Criteria
- Treat an absent pretrend rejection as insufficient proof of parallel trends.
- Do not pool incompatible cohort effects without explicit weights.
- Cluster uncertainty at the intervention assignment level where justified.
- A failed identification review blocks an unqualified causal conclusion.

## Contrast Case
An outcome rise in the treated region is not the effect if the comparison region rises equally.

## Dataset Scope
The attached starter files are prototype inputs. Expected starter answers are visible. They are not independent holdouts.

Full target: Use a synthetic region-by-month panel with known intervention timing and configurable confounding. Real administrative data requires separate authorization and identification review.

Complete target contracts and expert review remain pending. Public observations are not historical release vintages.

## Tools
pandas, statsmodels, Plotly

Use an available model appropriate to the task. Record provider, model identifier, version or retrieval date, settings, cost, and latency. Use deterministic code for calculations. Models may extract, retrieve, or explain evidence. They must not invent missing records, override approval rules, or claim professional authority.

## Constraints
The two-period fixture checks arithmetic, not a valid estimator for every staggered design.

- No fabricated records, source citations, professional endorsements, or expert-adjudication claims.
- No future-data leakage into dated decisions, model fitting, or historical valuation inputs.
- No unauthorized payments, account restrictions, regulatory filings, or production changes.
- No automated appraisal certification, tax-compliance determination, or unsupported fraud accusation.
- No concealed uncertainty or silently invented inputs required for a numerical conclusion.

## Scoring
- Correct domain calculations and justified methods with documented assumptions. Weight: 30.
- Source lineage, temporal integrity, rights checks, and explicit synthetic-data labels. Weight: 25.
- A working application with review, export, error handling, and reproducible state. Weight: 20.
- Adversarial cases, uncertainty, missing-data behavior, and sensitivity analysis. Weight: 15.
- Evidence-linked conclusions that distinguish facts, assumptions, and unresolved judgments. Weight: 10.
Score each dimension from zero to two. Weighted score = sum(weight * score / 2). A failed safety gate blocks acceptance.