run_modelrank#

pharmpy.tools.run_modelrank(models, results, ref_model, strictness='minimization_successful or (rounding_errors and sigdigs >= 0.1)', rank_type='ofv', alpha=0.05, search_space=None, E=None, parameter_uncertainty_method=None, **kwargs)#

Run ModelRank tool.

Parameters:

  • models (list[Model]) – Models to rank

  • results (list[ModelfitResults]) – Modelfit results to rank on

  • ref_model (Model) – Model to compare to

  • strictness (str or None) – Strictness criteria

  • rank_type (str) – Which ranking type should be used. Supported types are OFV, LRT, AIC, BIC (mixed, IIV, random), and mBIC (mixed, IIV, random). Default is OFV.

  • alpha (float) – Cutoff p-value that is considered significant in likelihood ratio test. Default is None

  • search_space (str, ModelFeatures or None) – Search space to test. Either as a string or a ModelFeatures object.

  • E (float) – Expected number of predictors (used for mBIC). Must be set when using mBIC. Tuple if mBIC for IIV (both diagonals and off-diagonals)

  • parameter_uncertainty_method ({‘SANDWICH’, ‘SMAT’, ‘RMAT’, ‘EFIM’} or None) – Parameter uncertainty method. Will be used in ranking models if strictness includes parameter uncertaint

  • kwargs – Arguments to pass to tool

Returns:

ModelRankResults – ModelRank tool result object

On this page