Generate a synthetic PMX dataset from a trial summary

Draws new subjects from a synpmx_pca_summarize() trial summary. This stage reads no patient data: its arguments are the model and a subject count, so everything the synthetic dataset is built from is visible in the model itself.

Usage

synpmx_pca_generate(trial_summary, n_subjects = NULL, seed = NULL)

Arguments

trial_summary: A pmx_trial_summary from synpmx_pca_summarize(), or a dataset generated from one.
n_subjects: Number of synthetic subjects. Defaults to the number the model was fitted on.
seed: Generation seed.

Value

A data frame in the source's shape, carrying the trial summary as an attribute.

Details

Each generated subject is assigned an arm, keeping each arm's share of the cohort. Its scores are that arm's mean plus a fresh residual, its dose schedule is drawn from the arm's planned schedule and the pooled dose-change rates, and it attends each visit its arm has at the attendance rate of that visit's endpoint. No individual's schedule and no individual's visit set exists in the model to be copied.

Examples

data <- pmx_simulated_fixture(60)
roles <- pmx_roles(
  id = "ID", time = "TIME", nominal_time = "NTIME", dv = "DV", amt = "AMT",
  evid = "EVID", cmt = "CMT", dvid = "DVID", mdv = "MDV"
)
trial_summary <- synpmx_pca_summarize(data, roles)
synthetic <- synpmx_pca_generate(trial_summary, seed = 1)
nrow(synthetic) > 0
#> [1] TRUE