Open problems · P1
Identifying personal parameters from three to five visits
Cohorts measure blood every two to four years. What can be identified about a person's own parameters (position, rate of drift, sensitivity) from so few points, and how should shrinkage toward the population be designed so it is honest about what one person's data can say?
Why it matters
The central PFL hypothesis is that a shared landscape plus a few personal parameters describes how each person ages. If those parameters cannot be identified from realistic data, the hypothesis cannot be tested, and the clinical use case fails.
What exists
- Within-person trajectories for 13 physical measures in HRS (aggregate tables on Ageing).
- Population hazard curves for 238 biomarkers (Biomarkers).
- Requested: repeated blood in HRS, CLSA, ARIC and Framingham.
A first step
Simulate people from the published HRS drift and tracking statistics, sample them on a 4-year schedule, and measure how well a hierarchical model recovers each person's rate of drift.