penAFT.trace: Print trace plot for the regularized Gehan estimator fit...

In penAFT: Fit the Regularized Gehan Estimator with Elastic Net and Sparse Group Lasso Penalties

Print trace plot for the regularized Gehan estimator fit using penAFT or penAFT.cv

Description

Print the trace plot for the regularized Gehan estimator.

Usage

penAFT.trace(fit, groupNames=NULL)

Arguments

fit	A fitted model from penAFT or penAFT.cv.
groupNames	A list of groupnames to be printed when fit used penalty "SG".

Details

The function penAFT.trace returns a trace plot for a fitted model obtained from either penAFT or penAFT.cv. If the model is fit using the sparse group-lasso penalty, you may provide names for the groups (in order of the integer values specifying the groups). This feature may not be desired if there are a large number of groups, however. The vertical blue line indicates the tuning parameter which minimized cross-validated linear predictor scores and the vertical black line indicates the tuning parameter minimizing the cross-validated Gehan loss according to the one-standard error rule.

Value

No return value; prints a trace plot as described in Details.

Examples

# --------------------------------------
# Generate data  
# --------------------------------------
set.seed(1)
genData <- genSurvData(n = 50, p = 50, s = 10, mag = 2, cens.quant = 0.6)
X <- genData$X
logY <- genData$logY
delta <- genData$status


# ------------------------------------------
# Fit elastic net penalized estimator with CV
# -------------------------------------------
fit.cv <- penAFT.cv(X = X, logY = logY, delta = delta,
                   nlambda = 10,
                   penalty = "EN",
                   alpha = 1, nfolds = 5)

# -- print plot
penAFT.trace(fit.cv)


  # ------------------------------------------
  # Fit sparse group-lasso estimator with CV
  # -------------------------------------------
  groups <- rep(1:5, length=10)
  fit.sg.cv <- penAFT.cv(X = X, logY = logY, delta = delta,
                    nlambda = 50, groups = groups,
                    penalty = "SG", tol.rel= 1e-5,
                    alpha = 0, nfolds = 5)
  
  penAFT.trace(fit.sg.cv, groupNames = paste("Group", 1:5, sep="-"))

penAFT documentation built on April 18, 2023, 9:10 a.m.