# Contract (P0)
- Question: is the memo's claim that the OLS slope of y on x is significant at 5% supported when the SE accounts for serial correlation?
- Estimand: beta in the linear projection y_t = a + b x_t + u_t, OLS on all 400 months. Meaningful under stationarity/weak dependence of (x_t,u_t); inference via Newey–West (Bartlett) HAC SE with justified bandwidth; normal reference distribution.
- Outputs: out/results.json (task_id, beta_hat, alpha_hat, se_iid, rho_resid, se_hac, hac_lags, t_hac, p_hac, beta_ci_low, beta_ci_high, memo_claim_supported, n_analysed, feasibility, feasibility_reason, data_issues, n_input_rows, random_seed), out/report.md (6 sections), out/analysis.py, out/claims.json (incl. memo-claim restatement).
- Information rules: only data/series.csv and data/memo.md; memo numbers only for comparison.
- Invalidating: missing/garbled data, gaps in months, non-reproducible script, relying on memo numbers.