sample_parameters_from_covariance_matrix#

pharmpy.modeling.sample_parameters_from_covariance_matrix(model, parameter_estimates, covariance_matrix, force_posdef_samples=None, force_posdef_covmatrix=False, n=1, seed=None)[source]#

Sample parameter vectors using the covariance matrix

If parameters is not provided all estimated parameters will be used

Parameters:

model (Model) – Input model
parameter_estimates (pd.Series) – Parameter estimates to use as means in sampling
covariance_matrix (pd.DataFrame) – Parameter uncertainty covariance matrix
force_posdef_samples (int) – Set to how many iterations to do before forcing all samples to be positive definite. None is default and means never and 0 means always
force_posdef_covmatrix (bool) – Set to True to force the input covariance matrix to be positive definite
n (int) – Number of samples
seed (Generator) – Random number generator

Returns:

pd.DataFrame – A dataframe with one sample per row

Example

>>> from pharmpy.modeling import *
>>> from pharmpy.tools import load_example_modelfit_results
>>> model = load_example_model("pheno")
>>> results = load_example_modelfit_results("pheno")
>>> rng = create_rng(23)
>>> cov = results.covariance_matrix
>>> pe = results.parameter_estimates
>>> sample_parameters_from_covariance_matrix(model, pe, cov, n=3, seed=rng)
     POP_CL    POP_VC   COVAPGR    IIV_CL    IIV_VC     SIGMA
0  0.004887  1.000761  0.198184  0.034860  0.031391  0.013750
1  0.004631  1.024746  0.071056  0.031726  0.026824  0.012597
2  0.004631  0.991088  0.130841  0.027464  0.024589  0.013215