Package index • BJM

Fit sub-models

Fit the survival and longitudinal sub-models that dynamic prediction is built on.

Predict event risk and future biomarker values from the fitted sub-models.

predictRisk(): Dynamic prediction function for future event risk
predictLongitudinal(): Dynamic prediction function for future longitudinal outcomes
simulateTrajectory(): Simulate future trajectories from the backward joint model
checkBandcountConvergence(): Check whether bandcount1/bandcount2/bandcount3 are large enough
survivalTrans(): Build a survival-time transform basis from cut points

Fit the backward joint model when the event time is only known to lie between two visits.

fitIntervalBJM(): Fit the backward joint model to interval-censored event times

Impute missing longitudinal biomarker values and pool sub-model fits across imputations.

imputeLongitudinal(): Impute missing longitudinal biomarker values with a deep generative model
poolLongitudinalSub(): Pool longitudinal sub-model fits across multiple imputations with Rubin's rules

Visualize predictions, fitted sub-models, observed longitudinal trajectories, and observed event outcomes.

predictPlot(): Plot risk and future biomarker predictions across a horizon sweep
riskPlot(): Plot predicted risk across a sweep of landmark times
cmtPlot(): Plot conditional mean trajectories (CMT)
spaghettiPlot(): Plot individual longitudinal trajectories (spaghetti plot)
cifPlot(): Plot cumulative incidence functions
performancePlot(): Plot predictive performance across landmark times
calibrationPlot(): Plot calibration of dynamic risk predictions
plot(<longitudinalSub.BJM>): Plot a fitted longitudinal sub-model
plot(<survivalSub.BJM>): Plot a fitted survival sub-model
plot(<predictLongitudinal.BJM>): Plot predicted biomarker distributions
plot(<simulateTrajectory.BJM>): Plot simulated futures