1[2 {"id":"C1","text":"Weekday mean daily changes in the 10y yield do not differ significantly (ANOVA F=0.26, p=0.90) over 2000-2019.","results_keys":["anova_f_stat","anova_p_value","n_analysed"],"status":"supported","caveat":"Classical ANOVA assumes independence and equal variances; volatility clustering not modelled."},3 {"id":"C2","text":"The Monday contrast is +0.05 bp with 95% Welch CI [-0.32, 0.42] bp, i.e. not distinguishable from zero.","results_keys":["effect_estimate","effect_ci_low","effect_ci_high"],"status":"supported","caveat":"Welch interval ignores serial dependence and time-varying volatility."},4 {"id":"C3","text":"Weekday-specific means estimated on 2000-2009 do not improve out-of-sample RMSE over a zero-change forecast in 2010-2019 (4.884 vs 4.876 bp).","results_keys":["model_metric","baseline_metric","n_train","n_test"],"status":"supported","caveat":"No formal forecast-comparison test; difference is tiny."},5 {"id":"C4","text":"Overall conclusion: no detectable day-of-week effect.","results_keys":["conclusion"],"status":"supported","caveat":"Small effects (< ~0.4 bp/day) cannot be ruled out."}6]7