1[2 {"id":"C1","text":"The conventional 95% OLS interval does not attain nominal coverage: estimated coverage 0.844, below the 0.95 - 3 MCSE threshold of 0.935.","results_keys":["coverage_conventional","conventional_ci_undercovers","undercoverage_threshold","mcse_coverage_at_nominal"],"status":"supported","caveat":"Single fixed design, normal errors, R=2000; MCSE ~0.008."},3 {"id":"C2","text":"Undercoverage arises because the mean conventional SE (0.190) understates the exact SD of the slope estimator (0.268).","results_keys":["mean_se_conventional","sd_beta_exact"],"status":"supported","caveat":"Mechanism attributed to sigma rising with |x - xbar| (leverage); correlation from the audit."},4 {"id":"C3","text":"The HC1 interval improves coverage to 0.902 but still falls short of 95% at n=30.","results_keys":["coverage_hc1","mcse_coverage_hc1","mean_se_hc1"],"status":"supported","caveat":"Shortfall not formally tested with the 3-MCSE rule; descriptive."},5 {"id":"C4","text":"The Monte Carlo SD of the slope estimator (0.2696) agrees with the analytic exact SD (0.2680), validating the simulation.","results_keys":["sd_beta_mc","sd_beta_exact"],"status":"supported","caveat":"Agreement judged against approx MC SE of a sample SD (~0.004)."}6]7