1[2 {"id":"C1","text":"The seasonal AR model (sar_aic) has lower out-of-sample RMSE than seasonal-naive at h=1, 2 and 4 (relative improvements 0.230, 0.135, 0.187).","results_keys":["rmse_model_h1","rmse_snaive_h1","rmse_model_h2","rmse_snaive_h2","rmse_model_h4","rmse_snaive_h4","rel_improvement_rmse_h1","rel_improvement_rmse_h2","rel_improvement_rmse_h4"],"status":"supported","caveat":"Point estimates over 40 origins; no significance test at h=1,2; current vintage data."},3 {"id":"C2","text":"The model also has lower MAE than seasonal-naive at all three horizons.","results_keys":["mae_model_h1","mae_snaive_h1","mae_model_h2","mae_snaive_h2","mae_model_h4","mae_snaive_h4"],"status":"supported","caveat":"Point estimates only."},4 {"id":"C3","text":"At h=4 the mean squared-error loss differential (2.82, SE 1.70, rectangular HAC lag 3) is not statistically distinguishable from zero at the 5% level (DM=1.66, p=0.097).","results_keys":["loss_diff_mean_h4","loss_diff_se_h4","hac_lags_h4","hac_kernel_h4","dm_stat_h4","dm_pvalue_h4"],"status":"supported","caveat":"Conclusion sensitive to HAC kernel/lag: p ranges 0.04-0.18 across alternatives in sensitivity_h4; small-sample asymptotics."}5]6