EthiopiaET

falling (-21.8pp)

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

73.5%

d6 · 2026-08-19

Floor day

28.7%

d30 · 2026-09-12

30-day delta

-21.8pp

d30 vs d1 probability

30-day trajectory with 90% conformal band

model trajectory_v1

mean P(shutdown by day)90% conformal bandpeak day

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

day 150.5%

2026-08-14

week_of_year-0.024
month-0.022
block_rate_roll7_mean-0.015

day 744.3%

2026-08-20

week_of_year-0.024
month-0.022
block_rate_roll7_mean-0.015

day 1468.0%

2026-08-27

week_of_year-0.024
month-0.022
block_rate_roll7_mean-0.015

day 3028.7%

2026-09-12

week_of_year-0.024
month-0.022
block_rate_roll7_mean-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

Day	Date	P(shutdown)	90% interval
d1	2026-08-14	50.5%	[7%, 94%]
d7	2026-08-20	44.3%	[0%, 100%]
d14	2026-08-27	68.0%	[0%, 100%]
d30	2026-09-12	28.7%	[0%, 100%]

Spotlight countries