# Data audit (P1)

• File: data/design.csv, SHA-256 2f459c539378cbe54d830c18af466f2fa4230ce48a609aa6464f8334a0d36ecb
• 7 rows x 3 columns (parameter, value, description). No missing values, no duplicated parameters. No dates (simulation design only); all parameters dimensionless except sigma (unit of y).
• Values: n=30, rho=0.8, sigma=1.0, intercept=yes, initial_condition=stationary (y_1 ~ N(0, sigma^2/(1-rho^2))), n_replications=20000, seed=20261029.
• Consistency: |rho|<1 so the stationary distribution exists; sigma>0; n>=3.
• Anomalies: none. Note: n is the series length (y_1..y_n), so the regression has n-1=29 observations; the analytical formula is evaluated with n=30 per the task ("at the design values"); n-1 variant reported as supplementary.
• Information-availability: only input is design.csv; no violations.
• Exclusions: replications are only dropped if rho_hat is non-finite (ledger in out/exclusions.csv; 0 removed).