TVARResult._summary_table#

TVARResult._summary_table()[source]#

Build the structured summary.

Return type:

SummaryTable

Returns:

The observed-regime rows (own first lag in each regime) and metadata; the notes state regime-wise stationarity, the regime split and the switching covariance, the delay search when it ran, the external driver when there is one, the non-identification of the threshold under linearity, the regime-conditional reading of the propagation methods, and the absence of standard errors.

Example

>>> import numpy as np
>>> y = np.random.default_rng(0).standard_normal((200, 2))
>>> fixed = TVAR(y, order=1, transition_variable="y1", delay=1).fit()._summary_table()
>>> fixed.title, fixed.columns, len(fixed.rows), len(fixed.notes)
('TVAR(1, d=1) Results', ('equation', 'lower own L1', 'upper own L1'), 2, 5)
>>> fixed.notes[1].startswith("Regime split:")
True
>>> searched = TVAR(y, order=2, transition_variable="y1").fit()._summary_table()
>>> len(searched.notes), "selected by the same grid search" in searched.notes[2]
(6, True)