PakistanPK
climbing (+58.5pp)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
75.6%
d23 · 2026-07-29
Floor day
6.0%
d3 · 2026-07-09
30-day delta
+58.5pp
d30 vs d1 probability
30-day trajectory with 90% conformal band
model trajectory_v1mean P(shutdown by day)90% conformal bandpeak day
Drivers (SHAP top 3 from the 7-day model — same features feed all horizons)
day 111.7%
2026-07-07
- block_rate_roll14_mean-0.015
- risk_tier+0.013
- block_rate_lag1+0.013
day 723.4%
2026-07-13
- block_rate_roll14_mean-0.015
- risk_tier+0.013
- block_rate_lag1+0.013
day 1453.7%
2026-07-20
- block_rate_roll14_mean-0.015
- risk_tier+0.013
- block_rate_lag1+0.013
day 3070.1%
2026-08-05
- block_rate_roll14_mean-0.015
- risk_tier+0.013
- block_rate_lag1+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
| Day | Date | P(shutdown) | 90% interval |
|---|---|---|---|
| d1 | 2026-07-07 | 11.7% | [0%, 56%] |
| d7 | 2026-07-13 | 23.4% | [0%, 88%] |
| d14 | 2026-07-20 | 53.7% | [0%, 100%] |
| d30 | 2026-08-05 | 70.1% | [0%, 100%] |
Spotlight countries