Skip to contents

Estimate small oracle prediction function using Super Learner regression

Source: R/prebuilt_generators.R
generate_oracle_predictions_SL.Rd

When the oracle prediction function is a conditional mean, the small oracle prediction function can be estimated by regressing the large oracle prediction function on the small feature vector. This function performs this regression using Super Learner.

Usage

generate_oracle_predictions_SL(
  time,
  event,
  X,
  X_holdout,
  nuisance_preds,
  outcome,
  landmark_times,
  restriction_time,
  approx_times,
  SL.library = c("SL.mean", "SL.glm", "SL.earth", "SL.gam", "SL.ranger"),
  V = 5,
  indx
)

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.

X

n x p data.frame of observed covariate values

X_holdout

m x p data.frame of new observed covariate values at which to obtain m predictions for the estimated algorithm. Must have the same names and structure as X.

nuisance_preds

Named list of conditional survival function predictions with elements "S_hat", "S_hat_train", "G_hat", and "G_hat_train". This should match the output of conditional_surv_generator.

outcome

Outcome type, either "survival_probability" or "restricted_survival_time"

landmark_times

Numeric vector of length J1 giving landmark times at which to estimate VIM ("accuracy", "AUC", "Brier", "R-squared").

restriction_time

Maximum follow-up time for calculation of "survival_time_MSE". Essentially, this time should be chosen such that the conditional survival function is identified at this time for all covariate values X present in the data. Choosing the restriction time such that roughly 10% of individuals remain at-risk at that time has been shown to work reasonably well in simulations.

approx_times

Numeric vector of length J2 giving times at which to approximate integral appearing in the pseudo-outcomes

SL.library

Super Learner library

V

Number of cross-validation folds, to be passed to SuperLearner

indx

Numeric index of column(s) of X to be removed, i.e., not used in the oracle prediction function.

Value

A list containing elements f0_hat and f0_hat_train, the estimated small oracle prediction functions for X_holdout and X, respectively.