1[2 {"id": "C1", "text": "Over the 60 origins 2021-09 to 2026-08 (h=1), the ridge forecast has lower RMSE than the naive forecast (0.561 vs 0.883 percent/month), a relative improvement of about 36%.", "results_keys": ["rmse_ridge", "rmse_naive", "rel_improvement_vs_naive", "n_origins"], "status": "supported", "caveat": "Current-vintage data, publication lags ignored; 60 evaluation points."},3 {"id": "C2", "text": "The improvement of ridge over the naive forecast is statistically significant under a Diebold-Mariano test with squared loss (HLN-corrected, p about 0.001).", "results_keys": ["dm_stat_ridge_vs_naive", "dm_pvalue_ridge_vs_naive"], "status": "supported", "caveat": "DM test treats the ridge model's estimation error as part of the forecast; asymptotic approximation with n=60."},4 {"id": "C3", "text": "The ridge forecast does not beat the expanding historical mean: its RMSE is about 3.1% higher (0.561 vs 0.544) and its MAE is also higher (0.451 vs 0.429).", "results_keys": ["rmse_ridge", "rmse_mean", "rel_improvement_vs_mean", "mae_ridge", "mae_mean"], "status": "supported", "caveat": "Single fixed lambda=50; no tuning was permitted."},5 {"id": "C4", "text": "The RMSE difference between ridge and the expanding mean is not statistically distinguishable from zero (DM p about 0.59).", "results_keys": ["dm_stat_ridge_vs_mean", "dm_pvalue_ridge_vs_mean"], "status": "supported", "caveat": "Low power with 60 out-of-sample errors."}6]7