VS: Function to perform variable selection

In BayesSurvive: Bayesian Survival Models for High-Dimensional Data

Function to perform variable selection

Description

Perform variable selection using the 95 neighborhood criterion (SNC), median probability model (MPM) or Bayesian false discovery rate (FDR). Note that the Bayesian FDR only applies for each subgroup if there are subgroups.

Usage

VS(x, method = "FDR", threshold = NA, subgroup = 1)

Arguments

x	fitted object obtained with BayesSurvive, or a matrix/array, or a list consisting of matrices and arrays
method	variable selection method to choose from c("CI", "SNC", "MPM", "FDR"). Default is "FDR"
threshold	SNC threshold value (default 0.5) or the Bayesian expected false discovery rate threshold (default 0.05)
subgroup	index(es) of subgroup(s) for visualizing variable selection

Value

A boolean vector of selected (= TRUE) and rejected (= FALSE) variables for one group or a list for multiple groups

References

Lee KH, Chakraborty S, Sun J (2015). Survival prediction and variable selection with simultaneous shrinkage and grouping priors. Statistical Analysis and Data Mining, 8:114-127

Newton MA, Noueiry A, Sarkar D, Ahlquist P (2004). Detecting differential gene expression with a semiparametric hierarchical mixture method. Biostatistics, 5(2), 155-76

Examples

library("BayesSurvive")
set.seed(123)

# Load the example dataset
data("simData", package = "BayesSurvive")

dataset <- list(
  "X" = simData[[1]]$X,
  "t" = simData[[1]]$time,
  "di" = simData[[1]]$status
)

# Initial value: null model without covariates
initial <- list("gamma.ini" = rep(0, ncol(dataset$X)))
# Hyperparameters
hyperparPooled <- list(
  "c0"     = 2, # prior of baseline hazard
  "tau"    = 0.0375, # sd for coefficient prior
  "cb"     = 20, # sd for coefficient prior
  "pi.ga"  = 0.02, # prior variable selection probability for standard Cox models
  "a"      = -4, # hyperparameter in MRF prior
  "b"      = 0.1, # hyperparameter in MRF prior
  "G"      = simData$G # hyperparameter in MRF prior
)


# run Bayesian Cox with graph-structured priors
fit <- BayesSurvive(
  survObj = dataset, hyperpar = hyperparPooled,
  initial = initial, nIter = 50, burnin = 30
)
# show variable selection
VS(fit, method = "FDR")

BayesSurvive documentation built on April 3, 2025, 5:58 p.m.