SamBrillman
summary
Joint shared-parameter models for multiple longitudinal biomarkers and time-to-event data get implemented in rstanarm stan_jm and fit via Stan, handling multivariate mixed models, multi-level clustering and several association structures (current value, slope, integral, lag). Example uses 312 PBC patients with bilirubin and albumin and reports a 2.1-fold increase in hazard per unit rise in log bilirubin. The package returns posterior samples, fitted trajectories and dynamic survival predictions; proportional-hazards limits and AFT extensions remain active issues.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:47Joint modeling context
- 3:57Purpose and benefits
- 5:48Mathematical formulation
- 10:26Software for joint models
- 11:52PBC dataset example
- 15:55Future directions
- 16:21Q&A
related talks
01h 24mBayesianism and Survival Analysis
34mFast, beautiful and easy Bayesian modelling: Can you have it all?
01h 27mFlexible Statistical Modeling
17mPrecision Medicine for All: Using Tidymodels to Validate PRS in Brazil
14mProbabilistic Programming in Python
21mHazards on the Causal Path
24mStochastic Block Models with R: Statistically rigorous clustering
35mEasy Model Blending using Pseudo-Bayesian Model Averaging in Python