SurvivalDART: SurvivalDART

Description

Fits an Accelerated Failure Time (AFT) model using the Dirichlet splitting prior (DART), which induces structural sparsity through a Beta–Dirichlet hierarchy on splitting probabilities.

Usage

SurvivalDART(
  time,
  status,
  X_train,
  X_test = NULL,
  timescale = "time",
  number_of_trees = 200,
  a_dirichlet = 0.5,
  b_dirichlet = 1,
  rho_dirichlet = NULL,
  k = 2,
  N_post = 1000,
  N_burn = 1000,
  verbose = TRUE,
  ...
)

Value

A fitted AFT-DART model object.

Arguments

time: Outcome vector of right-censored (non-negative) survival times.
status: Event indicator (1 = event, 0 = censored).
X_train: Design matrix for training data.
X_test: Optional test matrix. If NULL, predictions are computed at the column means of X_train.
timescale: Either "time" (log-transform internally) or "log" (already log-transformed).
number_of_trees: Number of trees in the ensemble. Default is 200.
a_dirichlet, b_dirichlet: Beta hyperparameters controlling sparsity in the Dirichlet splitting rule.
rho_dirichlet: Expected number of active predictors. If NULL, defaults to the number of covariates in X_train.
k: Scaling constant used to calibrate the prior variance of the step heights.
N_post: Number of posterior samples to store.
N_burn: Number of burn-in iterations.
verbose: Logical; print sampling progress.
...: Additional arguments passed to ShrinkageTrees to override default hyperparameters.

Details

This function provides a survival-specific wrapper for DART under an AFT formulation for right-censored outcomes.

Structural regularisation is induced through a Dirichlet prior on splitting probabilities, encouraging sparse feature usage in high-dimensional settings.

Users requiring alternative shrinkage priors on the leaf parameters (e.g., Horseshoe or half-Cauchy priors) should use ShrinkageTrees directly.