1# Verification checks (P4)23- Clean re-run of analysis.py: results.json byte-identical (cmp) -> PASS4- n_sub1+n_sub2==n_analysed==240 -> PASS5- exclusions sum: 319-sum(removed)==240 -> PASS6- effect_se>0 -> PASS7- CI contains estimate -> PASS8- CI symmetric 1.96*se -> PASS9- model_metric<baseline_metric -> PASS10- R2 identity 1-(model/baseline)^2 -> PASS11- sample_end<=2019-10-01 and start<end -> PASS12- chow p consistent with stability_conclusion -> PASS13- conclusion consistent with CI -> PASS14- table1 matches results -> PASS15- magnitude plausible (Okun beta ~ -0.2..-0.5 in literature-scale) -> PASS16- chow df == (2,236) -> PASS17