Estimate AUC VIM

Usage

vim_AUC(
  time,
  event,
  approx_times,
  landmark_times,
  f_hat,
  fs_hat,
  S_hat,
  G_hat,
  cf_folds,
  sample_split,
  ss_folds,
  robust = TRUE,
  scale_est = FALSE,
  alpha = 0.05
)

Arguments

time: n x 1 numeric vector of observed follow-up times If there is censoring, these are the minimum of the event and censoring times.
event: n x 1 numeric vector of status indicators of whether an event was observed. Defaults to a vector of 1s, i.e. no censoring.
approx_times: Numeric vector of length J1 giving times at which to approximate integrals.
landmark_times: Numeric vector of length J2 giving times at which to estimate AUC
f_hat: Full oracle predictions (n x J1 matrix)
fs_hat: Residual oracle predictions (n x J1 matrix)
S_hat: Estimates of conditional event time survival function (n x J2 matrix)
G_hat: Estimate of conditional censoring time survival function (n x J2 matrix)
cf_folds: Numeric vector of length n giving cross-fitting folds
sample_split: Logical indicating whether or not to sample split
ss_folds: Numeric vector of length n giving sample-splitting folds
robust: Logical, whether or not to use the doubly-robust debiasing approach. This option is meant for illustration purposes only — it should be left as TRUE.
scale_est: Logical, whether or not to force the VIM estimate to be nonnegative
alpha: The level at which to compute confidence intervals and hypothesis tests. Defaults to 0.05

Value

A data frame giving results, with the following columns:

landmark_time: Time at which AUC is evaluated.
est: VIM point estimate.
var_est: Estimated variance of the VIM estimate.
cil: Lower bound of the VIM confidence interval.
ciu: Upper bound of the VIM confidence interval.
cil_1sided: Lower bound of a one-sided confidence interval.
p: p-value corresponding to a hypothesis test of null importance.
large_predictiveness: Estimated predictiveness of the large oracle prediction function.
small_predictiveness: Estimated predictiveness of the small oracle prediction function.
vim: VIM type.
large_feature_vector: Group of features available for the large oracle prediction function.
small_feature_vector: Group of features available for the small oracle prediction function.

Estimate AUC VIM

Usage

Arguments

Value

See also