FAVARResult._summary_table#

FAVARResult._summary_table()[source]#

Build the structured summary.

Return type:

SummaryTable

Returns:

One row per factor with its pre-cleaning variance share; the metadata carries the block sizes, the order, whether the cleaning ran, the mean observation fit and the factor VAR’s log-likelihood; the notes state the stability verdict, the structural hook, the rotation caveat and – when no slow series were declared – the contamination warning.

Example

>>> import numpy as np
>>> rng = np.random.default_rng(0)
>>> panel = rng.standard_normal((150, 10))
>>> names = tuple(f"s{i}" for i in range(10))
>>> raw = FAVAR(panel, rng.standard_normal((150, 1)), n_factors=1, order=1).fit()
>>> table = raw._summary_table()
>>> table.title, table.columns, len(table.rows), len(table.notes)
('FAVAR(1) Results', ('factor', 'variance share'), 1, 4)
>>> table.metadata[5]
('Slow-variable cleaning', 'no')
>>> clean = FAVAR(
...     panel, rng.standard_normal((150, 1)), n_factors=1, order=1,
...     panel_names=names, slow=names[:5],
... ).fit()
>>> len(clean._summary_table().notes), clean._summary_table().metadata[5]
(3, ('Slow-variable cleaning', 'yes'))