# Contract (P0)

Question. Is the memo's claim that the OLS slope of y on x (monthly, 400 obs) is significant at 5% supported once uncertainty accounts for serially correlated errors?

Estimand. β in the linear projection y_t = α + β x_t + u_t, OLS on full sample. Meaningful under stationarity / weak dependence of (x_t, u_t), E[x_t u_t]=0 (projection definition). Inference: Newey–West HAC SE (Bartlett kernel), bandwidth justified; with AR(1) errors ρ≈0.7 and persistent x, the iid SE understates uncertainty.

Outputs. out/results.json keys: task_id, beta_hat, alpha_hat, se_iid, rho_resid, se_hac, hac_lags (int), t_hac, p_hac, beta_ci_low, beta_ci_high, memo_claim_supported (bool), n_analysed, feasibility, feasibility_reason, data_issues (list of {issue,handling}), n_input_rows ({"series.csv": n}), random_seed. Also out/report.md (sections Question; Data and exclusions; Method; Results; Uncertainty; Limitations and open issues), out/analysis.py, out/claims.json (incl. a claim restating the memo conclusion with my verdict), out/exclusions.csv, state files.

Information rules. Only data/series.csv and data/memo.md. Memo numbers may be reproduced for comparison, not taken as given.

Invalidity / impossibility. Missing/corrupt data, gaps in months (would break HAC time ordering), non-numeric values; unjustified bandwidth; using memo numbers as inputs.