# Contract — F1-conf-04

Question: Summarise the property-sales export data/sales.csv and reconcile it to the provider's control totals in data/metadata.json. Estimands (over records read unambiguously): n_valid_records; sum and mean of price_cad; median of price_cad/area_m2; reconciliation of count and price sum to controls (n_unreadable = control − valid; implied unreadable sum = control sum − valid sum). Meaningful assumptions: control totals refer to the same population as the export; valid records are read as-is (no imputation). 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, random_seed); out/analytic.csv; out/exclusions.csv; out/report.md; out/analysis.py; out/claims.json; state files. Info rules: only data/ files; no reconstruction/imputation/guessing of unreadable records; feasibility "complete" only if all control records analysed, else "partial" with exact count and reason. Invalidity: imputing corrupted lines; mismatched counts in ledger; using external info.