voidly
← Atlas

BelarusBY

climbing (+33.8pp)

30-day day-by-day shutdown-risk trajectory. Each point is a per-horizon XGBoost prediction (h=1..30) with a 90% conformal band derived from absolute LOCO residuals. Peak risk highlighted.

Peak day
77.6%
d19 · 2026-06-10
Floor day
1.5%
d1 · 2026-05-23
30-day delta
+33.8pp
d30 vs d1 probability

30-day trajectory with 90% conformal band

model trajectory_v1
0%25%50%75%100%d1d7d14d21d30peak d19 78%
mean P(shutdown by day)90% conformal bandpeak day

Drivers (SHAP top 3 from the 7-day model — same features feed all horizons)

day 11.5%
2026-05-23
  • critical_incident_7d-0.025
  • week_of_year+0.022
  • gdelt_unrest_30d+0.015
day 751.7%
2026-05-29
  • critical_incident_7d-0.025
  • week_of_year+0.022
  • gdelt_unrest_30d+0.015
day 1470.5%
2026-06-05
  • critical_incident_7d-0.025
  • week_of_year+0.022
  • gdelt_unrest_30d+0.015
day 3035.3%
2026-06-21
  • critical_incident_7d-0.025
  • week_of_year+0.022
  • gdelt_unrest_30d+0.015

Honest caveats

  • Per-horizon XGBoost stacked into a trajectory — not a true joint distribution.
  • Conformal bands are absolute-residual q90 from LOCO predictions on 18 eval countries.
  • Long horizons (h>=14d) have wider effective uncertainty than the conformal q90 suggests; the same lag features feed all horizons.
  • Bands floor at [0,1]; probabilities below 0.05 are at the model's noise floor.
  • SHAP attribution above is from the 7-day point model. The trajectory uses 30 separate per-horizon XGBoost models that share the same feature set; we have not yet computed per-horizon SHAP.

Sample horizons

DayDateP(shutdown)90% interval
d12026-05-231.5%[0%, 45%]
d72026-05-2951.7%[0%, 100%]
d142026-06-0570.5%[2%, 100%]
d302026-06-2135.3%[0%, 100%]
Spotlight countries