Rizopoulos 2016 - JMbayes Joint Models for Longitudinal and Time-to-Event Data
Routing Summary
This folder captures Rizopoulos (2016), The R package JMbayes for fitting joint models for longitudinal and time-to-event data using MCMC (J. Stat. Softw. 72(7)). 7 notes.
- Whole paper map and key result? → Rizopoulos 2016 - JMbayes Joint Models Overview
- Need the model equations and posterior? → Rizopoulos 2016 - Joint Model Specification and Posterior
- Need MCMC defaults, baseline-hazard splines, speed? → Rizopoulos 2016 - Implementation MCMC and Baseline Hazard
- Need the PBC fits and Student-t/binary/censored extensions? → Rizopoulos 2016 - PBC Basic Joint Model and Extended Longitudinal Outcomes
- Need value/slope/area/weighted/shared-RE association structures? → Rizopoulos 2016 - Association Structures and Transformation Functions
- Need dynamic predictions and Bayesian model averaging? → Rizopoulos 2016 - Dynamic Predictions and Bayesian Model Averaging
- Need AUC, dynamic C, prediction error, cross-validation? → Rizopoulos 2016 - Predictive Accuracy and Cross-Validation
Concept Map
| Concept | Note | Type | Depends On | Key Result |
|---|---|---|---|---|
| Joint model | Rizopoulos 2016 - Joint Model Specification and Posterior | definition | - | GLMM + relative-risk hazard linked by |
| Posterior factorisation | Rizopoulos 2016 - Joint Model Specification and Posterior | theorem | joint model | conditional independence given |
| Implementation | Rizopoulos 2016 - Implementation MCMC and Baseline Hazard | concept | posterior | RWM, penalised B-splines, 1 long chain |
| PBC results | Rizopoulos 2016 - PBC Basic Joint Model and Extended Longitudinal Outcomes | example | joint model | , HR 2.7 per doubling |
| Association structures | Rizopoulos 2016 - Association Structures and Transformation Functions | definition | joint model | value, slope, area, weights, shared RE |
| Dynamic prediction, BMA | Rizopoulos 2016 - Dynamic Predictions and Bayesian Model Averaging | theorem | joint model | $\pi_j(u |
| Accuracy | Rizopoulos 2016 - Predictive Accuracy and Cross-Validation | concept | prediction | AUC 0.854, C_dyn 0.791, CV AUC 0.851 |
Notes
- Rizopoulos 2016 - JMbayes Joint Models Overview — CONTAINS: question, key result, Source Map, E25 link.
- Rizopoulos 2016 - Joint Model Specification and Posterior — CONTAINS: Eqs. 1-7, priors, Gauss-Legendre cumulative hazard check, PyMC sketch.
- Rizopoulos 2016 - Implementation MCMC and Baseline Hazard — CONTAINS: design, defaults, convergence, speed, knots.
- Rizopoulos 2016 - PBC Basic Joint Model and Extended Longitudinal Outcomes — CONTAINS: PBC basic fit table and extended longitudinal outcomes.
- Rizopoulos 2016 - Association Structures and Transformation Functions — CONTAINS: five association structures, PBC estimate table, transformations.
- Rizopoulos 2016 - Dynamic Predictions and Bayesian Model Averaging — CONTAINS: dynamic predictions, Patient 2, BMA theorem.
- Rizopoulos 2016 - Predictive Accuracy and Cross-Validation — CONTAINS: AUC, dynamic C, PE estimators, accuracy table, validation recipe.
Sources
- Rizopoulos 2016 - The R Package JMbayes - Rizopoulos D (2016), J. Stat. Softw. 72(7), 46 pages.