1[2 {"id":"C1","text":"The classical one-way ANOVA does not reject equal weekday means of daily 10y yield changes (F=0.263, p=0.902).","results_keys":["anova_f_stat","anova_p_value","n_analysed"],"status":"supported","caveat":"Equal-variance assumption violated (Levene); Welch ANOVA and Kruskal-Wallis agree."},3 {"id":"C2","text":"The Monday contrast is +0.050 bp with 95% Welch CI [-0.319, 0.420] bp, i.e. not distinguishable from zero.","results_keys":["effect_estimate","effect_ci_low","effect_ci_high","effect_p_welch"],"status":"supported","caveat":"Effects smaller than ~0.4 bp/day cannot be excluded."},4 {"id":"C3","text":"Weekday-specific estimation-period means do not improve out-of-sample forecasts over a zero-change forecast (RMSE 4.884 vs 4.876 bp).","results_keys":["model_metric","baseline_metric","oos_rmse_ratio_model_over_baseline","oos_dm_p"],"status":"supported","caveat":"Single train/test split; the small deterioration is not significant at 5% (DM p=0.058)."},5 {"id":"C4","text":"Overall conclusion code: no_detectable_effect.","results_keys":["conclusion","anova_p_value"],"status":"supported","caveat":"Decided by the ANOVA at 5% as the task prescribes."},6 {"id":"C5","text":"The null result is robust to unequal variances, non-normality and autocorrelation in the Monday contrast SE.","results_keys":["sens_welch_anova_p","sens_kruskal_p","sens_levene_p","sens_hac_effect_p"],"status":"supported","caveat":"HAC with 5 lags only; no subperiod analysis."}7]8