# Data audit (P1)

File: data/design.csv

• SHA-256: 2f459c539378cbe54d830c18af466f2fa4230ce48a609aa6464f8334a0d36ecb (matches bundle.json)
• 7 data rows x 3 columns (parameter, value, description); header present.
• Parameters: n=30 (series length), 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.
• Missing values: none. Duplicate parameter names: none. Units: dimensionless (sigma in units of y).
• Date ranges: not applicable (simulated design, no dates).
• Anomalies: none. |rho|<1 so the stationary initial condition is well defined.
• Ambiguity noted: "n" in the formula -(1+3rho)/n is taken as the design series length n=30 (as TASK.md says "at the design values"); the regression uses n-1=29 observations. Sensitivity with 29 is mentioned in the report as a remark only.
• Information availability: the only input is the design; no rule violations possible.
• Exclusions: none expected; out/exclusions.csv records a check that drops non-finite rho_hat replications (expected 0 removed).