1[2 {"id":"C1","text":"The direct regression on lagged GDP growth and lagged LFS employment growth has lower RMSE and MAE than the naive forecast at h=1 (RMSE 0.743 vs 0.814 pp, ~9% improvement).","results_keys":["rmse_model_h1","rmse_naive_h1","mae_model_h1","mae_naive_h1","rel_improvement_rmse_h1"],"status":"supported","caveat":"DM test p=0.048 (borderline); gain only ~1% if COVID-era training pairs are kept."},3 {"id":"C2","text":"At h=3 the model reduces RMSE by about 21% relative to the naive forecast (0.902 vs 1.140 pp).","results_keys":["rmse_model_h3","rmse_naive_h3","rel_improvement_rmse_h3"],"status":"supported","caveat":"DM p=0.105, not significant at 5%; 48 overlapping origins, current vintage data."},4 {"id":"C3","text":"The improvement is attributable to the timely employment indicator.","results_keys":[],"status":"not_evaluated","caveat":"No GDP-only AR benchmark was estimated; part of the gain may come from mean reversion."},5 {"id":"C4","text":"At the last origin (2026-04) GDP is available through 2026-02 and employment through 2026-03 under the publication rules, and all forecasts respect these lags.","results_keys":["last_target_date_available","last_predictor_date_available"],"status":"supported","caveat":"Timing simulated on final-vintage data; not real-time first releases."}6]7