pharmr (version 2.2.0)

run_modelsearch: run_modelsearch

Description

Run Modelsearch tool. For more details, see :ref:modelsearch.

Usage

run_modelsearch(
  model,
  results,
  search_space,
  algorithm = "reduced_stepwise",
  iiv_strategy = "absorption_delay",
  rank_type = "bic",
  cutoff = NULL,
  strictness = "minimization_successful or (rounding_errors and sigdigs >= 0.1)",
  E = NULL,
  parameter_uncertainty_method = NULL,
  ...
)

Value

(ModelSearchResults) Modelsearch tool result object

Arguments

model

(Model) Pharmpy model

results

(ModelfitResults) Results for model

search_space

(str or ModelFeatures) Search space to test. Either as a string or a ModelFeatures object.

algorithm

(str) Algorithm to use.

iiv_strategy

(str) If/how IIV should be added to candidate models. Default is 'absorption_delay'.

rank_type

(str) Which ranking type should be used. Default is BIC.

cutoff

(numeric (optional)) Cutoff for which value of the ranking function that is considered significant. Default is NULL (all models will be ranked)

strictness

(str) Strictness criteria

E

(numeric or str (optional)) Expected number of predictors (used for mBIC). Must be set when using mBIC

parameter_uncertainty_method

(str (optional)) Parameter uncertainty method. Will be used in ranking models if strictness includes parameter uncertaint

...

Arguments to pass to tool

Examples

if (FALSE) {
model <- load_example_model("pheno")
res <- load_example_modelfit_results("pheno")
search_space <- 'ABSORPTION(ZO);PERIPHERALS(1)'
run_modelsearch(model=model, results=res, search_space=search_space, algorithm='exhaustive')
}