voidly

TurkmenistanTM

stable (+0.8pp)

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
17.4%
d11 · 2026-08-24
Floor day
1.8%
d2 · 2026-08-15
30-day delta
+0.8pp
d30 vs d1 probability

30-day trajectory with 90% conformal band

model trajectory_v1
0%25%50%75%d1d7d14d21d30peak d11 17%
mean P(shutdown by day)90% conformal bandpeak day

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

day 12.8%
2026-08-14
  • month-0.030
  • block_rate_roll7_mean-0.013
  • gdelt_conflict_tone_7d-0.012
day 72.5%
2026-08-20
  • month-0.030
  • block_rate_roll7_mean-0.013
  • gdelt_conflict_tone_7d-0.012
day 145.3%
2026-08-27
  • month-0.030
  • block_rate_roll7_mean-0.013
  • gdelt_conflict_tone_7d-0.012
day 303.6%
2026-09-12
  • month-0.030
  • block_rate_roll7_mean-0.013
  • gdelt_conflict_tone_7d-0.012

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
d12026-08-142.8%[0%, 47%]
d72026-08-202.5%[0%, 67%]
d142026-08-275.3%[0%, 74%]
d302026-09-123.6%[0%, 82%]
Spotlight countries