Generate distribution plots of between-occasion variability terms from Nonmem

Usage

NMplotBSV(
  data,
  regex.eta,
  names.eta = NULL,
  parameters = NULL,
  col.id = "ID",
  covs.num,
  covs.char,
  save = FALSE,
  show = TRUE,
  return.data = FALSE,
  title = NULL,
  file.mod,
  structure = "flat",
  use.phi,
  auto.map,
  keep.zeros = FALSE,
  keep.zeros.pairs = FALSE,
  ...
)

Arguments

data: A dataset - will be converted to data.frame so data.table is OK.
regex.eta: A regular expression defining the naming of the ETA's of interest. See `file.mod` too.
parameters: Character vector of model parameters to include. This will drop ETA's that are not associated with a parameter in this vector.
col.id: The name of the id column in data. Default is ID like Nonmem. This is not fully working if col.id is different from `ID`.
covs.num: Names of columns containing numerical covariates to plot the random effects against.
covs.char: Names of columns containing categorical covariates to plot the random effects against.
save: Save the generated plots?
return.data: If TRUE, the identified ETA's together with subject id and covariates will be returned in both wide and long format. If FALSE, you just get the plots.
file.mod: If used, parameter names that ETA's are associated with will be derived by looking at $PRED or $PK in the control stream. Essentially, it looks for where the ETA's are found and look for a parameter name to the left of a `=` sign. Alternatively, you can hard-code the ETA-parameter relationship using regex.eta.
fun.file: If saving plots, this function can be used to translate the file names. The inputs given to the function argument are "iov_pairs.png" and "iov_covs_n.png".
script: If saving the plots, a stamp to add. See ggstamp.