sample_individual_estimates#

pharmpy.modeling.sample_individual_estimates(model, individual_estimates, individual_estimates_covariance, parameters=None, samples_per_id=100, seed=None)[source]#

Sample individual estimates given their covariance.

Parameters:

model (Model) – Pharmpy model
individual_estimates (pd.DataFrame) – Individual estimates to use
individual_estimates_covariance (pd.DataFrame) – Uncertainty covariance of the individual estimates
parameters (list) – A list of a subset of individual parameters to sample. Default is None, which means all.
samples_per_id (int) – Number of samples per individual
seed (rng or int) – Random number generator or seed

Returns:

pd.DataFrame – Pool of samples in a DataFrame

Example

>>> from pharmpy.modeling import create_rng, load_example_model, sample_individual_estimates
>>> from pharmpy.tools import load_example_modelfit_results
>>> model = load_example_model("pheno")
>>> results = load_example_modelfit_results("pheno")
>>> rng = create_rng(23)
>>> ie = results.individual_estimates
>>> iec = results.individual_estimates_covariance
>>> sample_individual_estimates(model, ie, iec, samples_per_id=2, seed=rng)
             ETA_CL    ETA_VC
ID sample
1  0      -0.127941  0.037273
   1      -0.065492 -0.182851
2  0      -0.263323 -0.265849
   1      -0.295883 -0.060346
3  0      -0.012108  0.219967
...             ...       ...
57 1      -0.034279 -0.040988
58 0      -0.187879 -0.143184
   1      -0.088845 -0.034655
59 0      -0.187779 -0.014214
   1      -0.019953 -0.151151

[118 rows x 2 columns]