cv.regsurv: Title

In jeroenhoogland/regsurv: Regularized parametric survival modeling

Title

Description

Title

Usage

cv.regsurv(
  object,
  prep,
  nfolds = 10,
  plot = FALSE,
  force.nnhazards = TRUE,
  print = FALSE,
  maxit = 100L,
  feastol = 1e-08,
  ...
)

Arguments

object	an object of class regsurv
prep	an object of class survprep
nfolds	number of cross-validation folds
plot	plots deviance versus log(lambda) if TRUE
force.nnhazards	the force.nnhazards status as used for regsurv() is always maintained during cross-validation, but the force.nnhazards parameter in cv.regsurv() allows to force these constraints on the whole sample (TRUE) (as opposed to just the in-fold cases (when FALSE)).
print	prints progress if TRUE
maxit	(integer) maxit the maximum number of iterations for the (ecos) solver, default 100L
feastol	the tolerance on the primal and dual residual, default 1e-8
...	other parameters (only used by internal functions)

Value

oosll	out-of-sample log-likelihood
cvm	mean oosll per lambda
cvse	se of oosll per lambda
msdr	mean squared deviance residuals per lambda
lambda.grid	sed grid of lambda values (taken to be the same as those used to fit to regsurv model)
lambda.min	lambda value minimizing the (out-of-sample) deviance
lambda.min.index	index for the value of lambda minimizing the (out-of-sample) deviance

jeroenhoogland/regsurv documentation built on March 20, 2023, 3:37 a.m.