gemmr.sample_size.linear_model.fit_linear_model

gemmr.sample_size.linear_model.fit_linear_model(criterion, model, estr=None, tag=None, target_power=0.9, target_error=0.1, data_home=None, include_pc_var_decay_constants=None, include_latent_explained_vars=None)

Fit a linear model to outcome data.

Parameters:

• criterion (str) –

  Can be:

  • 'combined'

  • 'power'

  • 'association_strength'

  • 'weight'

  • 'score'

  • 'loading'

  • 'crossloading'

• model (str) – ‘cca’ or ‘pls’

• estr (None or sklearn-style estimator instance) – if not None must be compatible with model

• tag (str or None) – further specifies the outcome data file, cf. data.load_outcomes()

• target_error (float between 0 and 1) – target error level

• target_power (float between 0 and 1) – target power level

• data_home (None or str) – path where outcome data are stored, None indicates default path

• include_pc_var_decay_constants (bool) – whether to include a predictor for the principal component spectrum decay constant in the linear model

• include_latent_explained_vars (bool) – whether to include a predictor for the latent explained variance in the linear model

Returns:

lm – fitted model

Return type:

sklearn.LinearRegression instance