voidly

TurkeyTR

climbing (+42.9pp)

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
72.4%
d23 · 2026-06-14
Floor day
8.1%
d1 · 2026-05-23
30-day delta
+42.9pp
d30 vs d1 probability

30-day trajectory with 90% conformal band

model trajectory_v1
0%25%50%75%100%d1d7d14d21d30peak d23 72%
mean P(shutdown by day)90% conformal bandpeak day

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

day 18.1%
2026-05-23
  • block_rate_roll30_mean-0.094
  • week_of_year+0.087
  • incident_count_7d-0.026
day 740.8%
2026-05-29
  • block_rate_roll30_mean-0.094
  • week_of_year+0.087
  • incident_count_7d-0.026
day 1456.3%
2026-06-05
  • block_rate_roll30_mean-0.094
  • week_of_year+0.087
  • incident_count_7d-0.026
day 3050.9%
2026-06-21
  • block_rate_roll30_mean-0.094
  • week_of_year+0.087
  • incident_count_7d-0.026

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
d12026-05-238.1%[0%, 52%]
d72026-05-2940.8%[0%, 100%]
d142026-06-0556.3%[0%, 100%]
d302026-06-2150.9%[0%, 100%]
Spotlight countries