voidly
← Atlas

ChinaCN

climbing (+5.2pp)

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
17.7%
d15 · 2026-08-28
Floor day
1.3%
d5 · 2026-08-18
30-day delta
+5.2pp
d30 vs d1 probability

30-day trajectory with 90% conformal band

model trajectory_v1
0%25%50%75%d1d7d14d21d30peak d15 18%
mean P(shutdown by day)90% conformal bandpeak day

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

day 11.8%
2026-08-14
  • week_of_year-0.039
  • month-0.033
  • risk_tier-0.024
day 73.8%
2026-08-20
  • week_of_year-0.039
  • month-0.033
  • risk_tier-0.024
day 1413.5%
2026-08-27
  • week_of_year-0.039
  • month-0.033
  • risk_tier-0.024
day 307.0%
2026-09-12
  • week_of_year-0.039
  • month-0.033
  • risk_tier-0.024

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
d12026-08-141.8%[0%, 46%]
d72026-08-203.8%[0%, 69%]
d142026-08-2713.5%[0%, 82%]
d302026-09-127.0%[0%, 85%]
Spotlight countries