{"generated_at":"2026-10-23T06:10:13.404292+00:00","honest_caveats":["Performance estimate is the rolling-origin backtest (AUC ~0.82, PR-AUC ~0.29) \u2014 NOT train-set metrics. The deployed model is retrained on all data, so its train-set scores are meaningless.","Uses only OONI measurement signals (anomaly/confirmed rates, volumes, trajectories). No political-event / election / protest features yet \u2014 that is the next obvious lift.","Cross-country transfer (LOCO) is weaker for persistently-censored countries (SA/AE/EG/KZ AUC ~0.6) \u2014 the model needs some history of a country to forecast it well.","PR-AUC ~0.29 at a 10% base rate is a useful early-warning signal (precision ~0.31, recall ~0.42 at the F1-optimal threshold) \u2014 not a precise oracle."],"honest_validation":{"loco_auc_mean":0.7289645246691207,"loco_auc_median":0.743137122725092,"method":"strict temporal rolling-origin CV, 9 folds, on independent OONI ground truth \u2014 see honest-forecast-backtest.py","rolling_origin_climatology":{"auc":0.7537916783800023,"brier":null,"f1":0.33221273135329393,"pr_auc":0.22002138323237483,"precision":0.2181025596454393,"recall":0.7199928990136228},"rolling_origin_model":{"auc":0.8142401113434197,"brier":0.16361144391757726,"f1":0.3456844696488679,"pr_auc":0.29513154794839225,"precision":0.30416148632272033,"recall":0.4150551584468142},"rolling_origin_persistence":{"auc":0.594706596551582,"brier":null,"f1":0.2772239927784269,"pr_auc":0.16256085712393636,"precision":0.2788977685518728,"recall":0.2757054660016881}},"horizon_days":7,"label_definition":{"baseline_window":28,"event_floor":0.08,"event_std_floor":0.01,"event_vol_min":50,"event_z":2.0,"horizon_days":7,"min_vol":30},"loaded":true,"method":"independent OONI aggregation-API ground truth, country-relative z-score event label, strict temporal rolling-origin CV \u2014 see scripts/honest-forecast-backtest.py","model_version":"honest_forecast_v1","n_countries":58,"recommended_threshold":0.625,"schema":"voidly-honest-forecast-info/v1"}