bootstrap_mfmc_estimator

pyapprox.control_variate_monte_carlo.bootstrap_mfmc_estimator(values, weights, nbootstraps=10, verbose=True, acv_modification=True)

Boostrap the approximate MFMC estimate of the mean of high-fidelity data with low-fidelity models with unknown means

Parameters

valueslist (nmodels)

The evaluations of each information source seperated in form necessary for control variate estimators. Each entry of the list contains

values0np.ndarray (num_samples_i0,num_qoi)

Evaluations of each model used to compute the estimator \(Q_{i,N}\) of

values1: np.ndarray (num_samples_i1,num_qoi)

Evaluations used compute the approximate mean \(\mu_{i,r_iN}\) of the low fidelity models.

weightsnp.ndarray (nmodels-1)

The control variate weights

nbootstrapsinteger

The number of boostraps used to compute estimator variance

verbose:

If True print the estimator mean and +/- 2 standard deviation interval

Returns

bootstrap_meanfloat

The bootstrap estimate of the estimator mean

bootstrap_variancefloat

The bootstrap estimate of the estimator variance