1# Verification checks (P4)2Clean re-run: deleted results.json/table1.csv/figure1.png/exclusions.csv, re-ran analysis.py → results.json byte-identical (cmp). PASS3- n_train+n_test==n_analysed (2500+2501=5001): PASS4- table1 n sums to n_analysed: PASS5- exclusions: 5217 − (215+1) = 5001 = n_analysed: PASS6- n_missing_rows independently recomputed (pandas NaN count = 215): PASS7- CI contains estimate: PASS8- All SEs positive: PASS9- Monday contrast recomputed from table1 (n-weighted Tue–Fri mean): PASS10- ANOVA p in [0,1]: PASS11- conclusion consistent with ANOVA p ≥ 0.05 rule: PASS12- RMSE magnitude plausible vs. sd of daily changes (~5 bp): PASS13- Baseline RMSE independently recomputed: PASS14- Information rule: first evaluation change dated 2010-01-04; weekday means use train only (code: groupby on train): PASS15- Figure visually inspected: five weekday means with 95% bars, all covering zero: PASS16- Assumption check: Levene test rejects equal variances (p≈1e-7) → classical ANOVA assumption violated; Welch ANOVA (p≈0.90) and Kruskal–Wallis (p≈0.52) agree with classical result. Reported, not a failure of the computation.17