Skip to contents

Optimal Survival Cutpoint Search for Time-Varying Counting Process Data

Source: R/km_cutpoints.R
km_cutpoint_time_varying.Rd

Evaluates candidate biomarker cutpoints on start-stop survival data (`Surv(tstart, tstop, status)`), clustering by subject ID to avoid data leakage.

Usage

km_cutpoint_time_varying(
  data,
  tstart_col,
  tstop_col,
  status_col,
  var_col,
  id_col,
  min_prop = 0.1,
  n_cuts = 30
)

Arguments

data

Long-format data frame with counting process layout.

tstart_col

Character name of interval start time.

tstop_col

Character name of interval stop time.

status_col

Character name of event status column.

var_col

Character name of continuous biomarker column.

id_col

Character name of subject ID column.

min_prop

Minimum proportion of total records in each subgroup (default 0.10).

n_cuts

Number of candidate cutpoints (default 30).

Value

A list with class `"cbe_km_cutpoint"` containing optimal cutpoint and candidate statistics.