survivalsurrogate (version 1.1)

estimate_R: Estimates the proportion of the treatment effect explained

Description

Estimates the proportion of the treatment effect on the censored primary outcome that is explained by the longitudinal surrogate marker, given the estimated influence functions for the overall treatment effect (delta) and the residual treatment effect (delta_s).

Usage

estimate_R(delta_if, delta_s_if, delta = NULL, delta_s = NULL, se_type = "asymptotic",
n_boot = NULL, alpha = 0.05)

Value

A dataframe with the following components:

estimand

Indicates what estimand is estimated, including the treatment effect (delta), residual treatment effect (delta.s), and proportion of treatment effect explained (R) by the longitudinal surrogate marker.

estimate

Estimates of the treatment effect (delta), residual treatment effect (delta.s), and proportion of treatment effect explained (R) by the longitudinal surrogate marker.

se

Estimated standard errors.

ci_l

Lower bound of the confidence intervals.

ci_h

Upper bound of the confidence intervals.

Arguments

delta_if

Influence function estimates for delta.

delta_s_if

Influence function estimates for delta_s.

delta

(Optional) Defaults to the mean of the delta influence function estimates.

delta_s

(Optional) Defaults to the mean of the delta_s influence function estimates.

se_type

Type of standard error estimation, choices are "asymptotic" or "bootstrap".

n_boot

(Optional unless se_type = "bootstrap") Number of boostrap samples.

alpha

(Optional) Alpha level used for confidence interval. If not provided, this is set to 0.05.

References

Agniel D and Parast L (2025). "Robust Evaluation of Longitudinal Surrogate Markers with Censored Data." Journal of the Royal Statistical Society: Series B; tools:::Rd_expr_doi("10.1093/jrsssb/qkae119").

Examples

data(exampledata)
names(exampledata)
library(glue)
library(rpart)
library(mlr3)
library(dplyr)
library(mlr3learners)

tt <- 5
t0 <- 4
yvars <- paste0('Y_', 0:tt)
lrnc <- glue('regr.rpart')
lrnb <- glue('classif.log_reg')

# \donttest{
p_deltahat <- plugin_delta(
    data = exampledata,
    folds = 'ff',
    id = 'ID',
    x = 'X_0',
    g = 'G_0',
    a = paste0('A_', 0:tt),
    y = yvars,
    s = paste0('S_', 0:t0),
    binary_lrnr = lrn(lrnb, predict_type = 'prob'),
    cont_lrnr = lrn(lrnc),
    truncate_e = 0.005,
    verbose = FALSE
  )

p_deltahat_s <- plugin_delta_s(
    data = exampledata,
    folds = 'ff',
    id = 'ID',
    x = 'X_0',
    g = 'G_0',
    a = paste0('A_', 0:tt),
    y = yvars,
    s = paste0('S_', 0:t0),
    binary_lrnr = lrn(lrnb, predict_type = 'prob'),
    cont_lrnr = lrn(lrnc),
    t0=t0,
    truncate_pi = 0.005,
    truncate_e = 0.005,
    verbose = FALSE
  )


estimate_R(p_deltahat$if_data[[1]]$eif,
p_deltahat_s$if_data[[1]]$eif)
# }