MarkovSwitchingARResult.regime_table#

MarkovSwitchingARResult.regime_table()[source]#

One row per regime: level, scale, and how long it lasts.

The complement to the coefficient table, which reads down parameters; this reads across regimes, which is the comparison the model exists to support. Autoregressive coefficients are left to params – there are p of them per regime and they do not fit a fixed-width row.

Return type:

SummaryTable

Returns:

A SummaryTable with one row per regime.

Example

>>> import numpy as np
>>> rng = np.random.default_rng(0)
>>> res = MarkovSwitchingAR(rng.standard_normal(300), order=1).fit(seed=0, n_init=2)
>>> table = res.regime_table()
>>> table.columns, len(table.rows)
(('regime', 'const', 'sigma2', 'ergodic', 'duration', 'share'), 2)
>>> table.metadata[0]
('Ordering', 'ascending intercept')