Panel Model

PanelSW2023

class sw2023.PanelSW2023(X, Y, firm_id, time_id, direction='mean', method='HMS', h=None, log_transform=True, standardize=True, bandwidth_method='silverman')[source]

Bases: object

4-Component Panel SFA with SW(2023) Rotation Transformation.

Parameters:

  • X ((n, p) inputs)

  • Y ((n, q) outputs)

  • firm_id ((n,) individual identifier (farm ID, etc.))

  • time_id ((n,) time identifier (year, etc.))

  • direction ('mean' | 'median' | array direction vector)

  • method ('SVKZ' | 'HMS')

  • h (bandwidth (None = Silverman))

  • log_transform (bool)

  • standardize (bool)

  • bandwidth_method ({'silverman', 'loocv', 'loocv_scalar'}) – Bandwidth selection method for pooled moment estimation.

__repr__()[source]

Return repr(self).

fit(verbose=True)[source]

Estimate the 4-component model.

summary()[source]

Print estimation results summary and return DataFrame.

results_dataframe()[source]

Return full results as a DataFrame.