1[2 {"id":"C1","text":"The seasonal AR model has lower out-of-sample RMSE and MAE than the seasonal-naive forecast at h=1, 2 and 4 (RMSE improvements of 23%, 14%, 19%).","results_keys":["rmse_snaive_h1","rmse_model_h1","rmse_snaive_h2","rmse_model_h2","rmse_snaive_h4","rmse_model_h4","mae_snaive_h1","mae_model_h1","mae_snaive_h2","mae_model_h2","mae_snaive_h4","mae_model_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."},3 {"id":"C2","text":"At h=4 the mean squared-error loss differential (2.83 sq. pp, favouring the model) is not statistically distinguishable from zero at 5% with a Bartlett HAC SE at lag 4 (DM=1.53, p=0.125).","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":"Fragile: longer HAC lags (6, 8) give DM > 1.96; small sample of 40 overlapping errors."},4 {"id":"C3","text":"The h=4 test conclusion is sensitive to the HAC truncation lag and kernel.","results_keys":["sens_bartlett_lag3_dm_stat_h4","sens_bartlett_lag6_dm_stat_h4","sens_bartlett_lag8_dm_stat_h4","sens_rect_lag3_dm_stat_h4","sens_hln_dm_stat_h4","sens_hln_dm_pvalue_h4_t39"],"status":"supported","caveat":"Sensitivity analysis only."}5]6