segreg.bootstrap.boot_param_dist

boot_param_dist(indep, dep, estimator, num_sims, resample_cases=False, seed=None, verbose=False, diagnoser=None, include_fixed_params=True)[source]

Computes bootstrap sampling distribution of model parameters.

Parameters

  • indep (array-like) – The independent data. Also called predictor, explanatory variable, regressor, or exogenous variable.

  • dep (array-like) – The dependent data. Also called response, regressand, or endogenous variable.

  • estimator (subclass of segreg.model.estimator.Estimator) –

  • num_sims (int) – Number of bootstrap simulations.

  • resample_cases (boolean, default False) – If True, the bootstrap will resample pairs with replacement from (indep, dep). See Section 6.2.4 in Davison and Hinkley, “Bootstrap Methods and their Application”.

  • seed (int) – Seed for random generator driving bootstrap simulations.

  • verbose (bool) –

  • diagnoser (function object taking params: estimator, params, indep, dep) – Used for diagnosing each bootstrap resample. This is currently a developmental feature.

  • include_fixed_params (bool) –

Returns

param sims – Returns bootstrap sampling distribution for the parameters: panel data where each row represents parameter estimates for a bootstrap sample. The columns are in the same order as given by the estimator input.

Return type

numpy array shape (num_sims, num_params)