voidly

EthiopiaET

stable (+0.4pp)

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
70.9%
d10 · 2026-06-18
Floor day
32.9%
d29 · 2026-07-07
30-day delta
+0.4pp
d30 vs d1 probability

30-day trajectory with 90% conformal band

model trajectory_v1
0%25%50%75%100%d1d7d14d21d30peak d10 71%
mean P(shutdown by day)90% conformal bandpeak day

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

day 136.7%
2026-06-09
  • gdelt_unrest_30d+0.046
  • week_of_year-0.029
  • block_rate_lag3+0.021
day 752.5%
2026-06-15
  • gdelt_unrest_30d+0.046
  • week_of_year-0.029
  • block_rate_lag3+0.021
day 1447.9%
2026-06-22
  • gdelt_unrest_30d+0.046
  • week_of_year-0.029
  • block_rate_lag3+0.021
day 3037.1%
2026-07-08
  • gdelt_unrest_30d+0.046
  • week_of_year-0.029
  • block_rate_lag3+0.021

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
d12026-06-0936.7%[0%, 81%]
d72026-06-1552.5%[0%, 100%]
d142026-06-2247.9%[0%, 100%]
d302026-07-0837.1%[0%, 100%]
Spotlight countries