# Contract (F1-conf-04)

Question: Summarise the provider's property-sales export (data/sales.csv) and reconcile it to provider control totals (data/metadata.json). Estimands: n of records readable unambiguously (n_valid_records = n_analysed); n_unreadable_records = control n − valid n; sum and mean price_cad over valid records; median of price_cad/area_m2 over valid records; implied unreadable price sum = control sum − valid sum. Meaningful assuming control totals describe the same table before export and that valid-record fields are correct as written. Outputs: out/results.json (task_id, n_records_control, n_valid_records, n_unreadable_records, n_analysed, sum_price_control, sum_price_valid, sum_price_unreadable_implied, mean_price_valid, median_price_per_m2_valid, feasibility, feasibility_reason, data_issues, n_input_rows{"sales.csv"}, random_seed); out/analytic.csv (n_analysed rows, 5 columns); out/exclusions.csv (chained ledger from data lines to n_analysed); out/report.md (6 sections); out/analysis.py; out/claims.json. Information rules: only data/ files; no reconstruction/imputation/guessing of unreadable records; feasibility "complete" only if every control record is analysed, else "partial" with exact count and reason. Invalidating conditions: imputing garbled rows; counting garbled rows as valid; using external data.