voidly

IranIR

stable (+0.2pp)

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
18.6%
d9 · 2026-06-24
Floor day
1.9%
d16 · 2026-07-01
30-day delta
+0.2pp
d30 vs d1 probability

30-day trajectory with 90% conformal band

model trajectory_v1
0%25%50%75%d1d7d14d21d30peak d9 19%
mean P(shutdown by day)90% conformal bandpeak day

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

day 14.6%
2026-06-16
  • block_rate_roll7_std-0.021
  • gdelt_conflict_tone_7d+0.014
  • week_of_year-0.013
day 77.1%
2026-06-22
  • block_rate_roll7_std-0.021
  • gdelt_conflict_tone_7d+0.014
  • week_of_year-0.013
day 143.7%
2026-06-29
  • block_rate_roll7_std-0.021
  • gdelt_conflict_tone_7d+0.014
  • week_of_year-0.013
day 304.8%
2026-07-15
  • block_rate_roll7_std-0.021
  • gdelt_conflict_tone_7d+0.014
  • week_of_year-0.013

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
d12026-06-164.6%[0%, 48%]
d72026-06-227.1%[0%, 72%]
d142026-06-293.7%[0%, 73%]
d302026-07-154.8%[0%, 83%]
Spotlight countries