chendun1995 and Claude Opus 5.5
ae58d0f7ca
D57: run-time trims (V3 skipped where physician plays no role, NCE without V3, landmark_m5 under V1 only)
...
Co-Authored-By: Claude Opus 5.5 <[email protected]
2026-10-28 16:54:42 -04:00
chendun1995 and Claude Opus 5.5
d369a16208
D55: EHR/registry-like world (noisy severity), non-random trial consent, trial analyses at 12/24/60 months
...
Co-Authored-By: Claude Opus 5.5 <[email protected]
2026-10-28 16:26:01 -04:00
chendun1995 and Claude Opus 5.5
5304c89316
D53: landmark on baseline covariates (month-5 version as landmark_m5), NCO-1 depends on PVI-free severity, physician-cluster bootstrap script
...
Co-Authored-By: Claude Opus 5.5 <[email protected]
2026-10-28 07:20:32 -04:00
chendun1995 and Claude Opus 5.5
cfc9da2c87
Speed-ups (shared PVI fit, clones once, case-control sampled deferred model); final scale 200 reps extending the draft (D51)
...
Co-Authored-By: Claude Opus 5.5 <[email protected]
2026-10-27 09:31:03 -04:00
chendun1995 and Claude Opus 5.5
383ef89c2a
Validation, NCE analysis, per-replicate oracle for landmark/naive, RCT truths script, resumable experiment runner
...
- W0 harm validation (40 reps): CCW and sequential ITT unbiased for their truths (|z| <= 0.7)
- PVI models fitted separately inside and after the window
- experiments.yml grid (8 worlds x 3 scenarios; designs x V1-V4; NCE; RCTs); smoke run clean
Co-Authored-By: Claude Opus 5.5 <[email protected]
2026-10-27 09:19:59 -04:00
chendun1995andClaude Opus 5.5