predrisk_bin_segments: Bin predicted risk for plotting

In bcjaeger/survival.calib: Tests of Calibration in the Survival Setting

Bin predicted risk for plotting

Description

Bin predicted risk for plotting

Usage

predrisk_bin_segments(
  x,
  event_time = NULL,
  event_status = NULL,
  pred_horizon = NULL,
  by_event = FALSE,
  bin_count = 100,
  bin_yintercept = 0,
  bin_length = 1
)

Arguments

x	a numeric vector of predicted risk values or an object of class scalib. If x is a scalib object, then the input arguments event_time, event_status, and pred_horizon should be left blank. (They are filled in using info stored in the scalib object).
event_time	(numeric vector) observed event times
event_status	(numeric vector) observed event status. The values of this vector should be 0 (event censored) and 1 (event observed).
pred_horizon	(numeric value) the time of risk prediction.
by_event	If TRUE, bins will be created for each event type, separately. If FALSE, bins are made in the standard fashion for histograms.
bin_count	(integer value) total count of bins for downstream plots
bin_yintercept	(numeric value) where, relative to the y-axis, the bins should originate from on downstream plots.
bin_length	(numeric value) the length of the bins on downstream plots.

Value

a data.frame object with values that can be plugged into standard plotting tools, e.g., ggplot2::ggplot() (see examples).

Examples

sc <- scalib(pred_risk = pbc_scalib$predrisk,
             pred_horizon = 2500,
             event_time = pbc_scalib$test$time,
             event_status = pbc_scalib$test$status)

pbins <- predrisk_bin_segments(sc)

print(pbins)

pbins <- predrisk_bin_segments(x = pbc_scalib$predrisk$prop_hazard,
                               event_time = pbc_scalib$test$time,
                               event_status = pbc_scalib$test$status)

bcjaeger/survival.calib documentation built on June 15, 2022, 7:47 a.m.