V5 Model Note
V5 is now the live structural release. This artifact preserves the final comparison against the retained V4.3 baseline and keeps the transition-adjusted and realized-risk layers auditable.
This page now preserves the final V5 promotion comparison. The planning rationale and next-step calibration work stay on the V5 roadmap.
Status
Live now
published in occupations with retained V4.3 comparison
Structural validation
2/2
BLS and family checks versus V4.3
Realized validation
3/4
scorable short-run checks retained in the audit trail
Task-mode blended
485
occupations with mode-aware exposure blend
Median fragility
0.0111
concentration-weighted task fragility
What Is Integrated
Structural layer
Latent posterior + task modeSource-calibrated latent exposure stays the anchor, while task-mode proxies derived from weighted O*NET tasks adjust the exposure layer where the task evidence is strong enough.
Concentration layer
Fragility + reallocationHampole-style concentration now enters as demand fragility and reallocation capacity. A few highly exposed core duties are treated differently from diffuse exposure spread across many smaller tasks.
Adaptation layer
Augmentation + hybrid mobilityHeterogeneous augmentation and hybrid mobility become a bounded adaptation buffer, creating a transition-adjusted risk view rather than directly overwriting structural risk.
Realized-risk layer
Short runThe realized-risk proxy remains conservative and separate, reflecting archetype-specific short-run caps, adoption timing, postings resistance, and offset room rather than raw structural overlap.
Governance
Promoted liveThe live V5 release remains auditable because the retained V4.3 baseline, sidecars, and promotion-comparison artifacts are still published separately.
Validation Snapshot
Structural checks
BLS rho
live -0.1905 → experimental -0.2455
Family rho
live -0.328 → experimental -0.5286
Realized-risk checks
Vacancy trend rho
-1
Hiring-net rho
0.5
Scorable, but still only a coarse 3-cluster diagnostic
Retrenchment rho
0
Current cluster series is too flat to score honestly
Employer-pressure rho
0.4762
positive coherence across archetypes, sample 432
Postings-support rho
-0.0896
negative coherence against direct occupation-level postings support, sample 93
Profile Mix
shielded
257
occupations
leveraged
23
occupations
retooling
126
occupations
balanced
156
occupations
Largest Changes
Structural risers
- Storekeeper (0.1625)
- Stock records clerk (0.1562)
- Tailor/Dressmaker (0.1077)
- Precision grinding machine setter-operator (0.0995)
- Keeper/Trainer in zoological, bird and aquatic parks (0.0987)
Transition fallers
- Telemarketer (-0.197)
- Clearing and forwarding agent (-0.1824)
- Shipping agent/Boarding officer (-0.1801)
- Enumerator/Market research interviewer (-0.174)
- Statistical clerk (-0.1662)
Highest realized risk
- Office clerk (including filing and copying) (0.2436)
- Securities clerk (0.2419)
- Enumerator/Market research interviewer (0.2391)
- Statistical clerk (0.2383)
- Postman (0.231)
Downloads