SDRreg.SuperLearner: Regression based on sequentially doubly robust transformation...

In QIU-Hongxiang-David/SDRsurv: Sequentially Doubly Robust Estimation of Survival Curve with Time-Varying Covariates

Regression based on sequentially doubly robust transformation of fitted survival and censoring probabilities using SuperLearner

Description

Apply doubly robust transformation on fitted survival and censoring probabilities in each time window and estimate P(T > t | T > truncation time, covariates available at truncation time) with SuperLearner::SuperLearner.

Usage

SDRreg.SuperLearner(
  covariates,
  follow.up.time,
  pred_event_censor.list,
  visit.times,
  tvals,
  truncation.index,
  id.var,
  time.var,
  event.var,
  Q.formula = ~.,
  Q.SuperLearner.control = list(family = gaussian(), SL.library = "SL.lm"),
  denom.survival.trunc = 0.001
)

Arguments

covariates	see SDRsurv
follow.up.time	see SDRsurv
pred_event_censor.list	list of pred_event_censor objects (see pred_event_censor). Each pred_event_censor object in the list corresponds to a time window in visit.times after truncation.index in increasing order.
visit.times	see SDRsurv
tvals	see SDRsurv. Must be sorted in ascending order.
truncation.index	see SDRsurv
id.var	see SDRsurv
time.var	see SDRsurv
event.var	see SDRsurv
Q.formula	formula to specify covariates being used for estimating P(T > t | T > visit.times[truncation.index], covariates available at visit.times[truncation.index]). Set to include intercept only (~ 0 or ~ -1) for marginal survival probability, which is simply the mean of pseudo-outcomes. Default is ~ ., which includes main effects of all available covariates up to (inclusive) the truncation.time.
Q.SuperLearner.control	see SDRsurv
denom.survival.trunc	see SDRsurv

Value

a list of mult_stage_survfit objects, each corresponding to a value in tvals

Warning

This function is designed to be called by other functions such as SDRsurv, therefore inputs are not thoroughly checked. Incorrect inputs may lead to errors with non-informative messages. The user may call this function if more flexibility is desired.

Custom learners

Custom learners may be specified by providing an element named SL.library in Q.SuperLearner.control.The user may refer to resources such as https://cran.r-project.org/web/packages/SuperLearner/vignettes/Guide-to-SuperLearner.html for a guide to create custom learners.

QIU-Hongxiang-David/SDRsurv documentation built on March 29, 2024, 8:41 a.m.