voidly

ChinaCN

stable (+2.3pp)

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
14.8%
d27 · 2026-07-16
Floor day
1.4%
d6 · 2026-06-25
30-day delta
+2.3pp
d30 vs d1 probability

30-day trajectory with 90% conformal band

model trajectory_v1
0%25%50%75%d1d7d14d21d30peak d27 15%
mean P(shutdown by day)90% conformal bandpeak day

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

day 14.0%
2026-06-20
  • block_rate_roll7_std-0.009
  • week_of_year-0.008
  • high_urgency_signals_7d-0.007
day 73.3%
2026-06-26
  • block_rate_roll7_std-0.009
  • week_of_year-0.008
  • high_urgency_signals_7d-0.007
day 145.8%
2026-07-03
  • block_rate_roll7_std-0.009
  • week_of_year-0.008
  • high_urgency_signals_7d-0.007
day 306.3%
2026-07-19
  • block_rate_roll7_std-0.009
  • week_of_year-0.008
  • high_urgency_signals_7d-0.007

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
d12026-06-204.0%[0%, 48%]
d72026-06-263.3%[0%, 68%]
d142026-07-035.8%[0%, 75%]
d302026-07-196.3%[0%, 85%]
Spotlight countries