Simulate attained-age survival data under a Cox PH model

Usage

sim_cox_age_data(
  n,
  entry_age,
  censor_age,
  beta,
  covariates,
  baseline_hazard = 0.004,
  baseline_hazard_by_age = NULL,
  target_avg_baseline_hazard = NULL
)

Arguments

n: Integer, number of subjects.
entry_age: Numeric scalar or length-n vector of entry ages.
censor_age: Numeric scalar or length-n vector of administrative censoring ages.
beta: Numeric vector of regression coefficients with length p.
covariates: Numeric matrix or data.frame with n rows and p columns.
baseline_hazard: Optional positive scalar specifying a constant baseline hazard over attained age. Must be NULL if baseline_hazard_by_age is provided.
baseline_hazard_by_age: Optional data.frame with columns age_lo, age_hi, hazard defining piecewise-constant baseline hazard over attained age intervals [age_lo, age_hi). Must be NULL if baseline_hazard is provided.
target_avg_baseline_hazard: Optional positive scalar. If provided, rescales the baseline hazard so the cohort-weighted average equals this value.

Value

data.frame with id, entry_age, censor_age, time, event and covariates.

Details

Exactly one of baseline_hazard or baseline_hazard_by_age must be specified (i.e. one must be non-NULL). Supplying both or neither results in an error.