MarkovSwitchingARResult.is_regimewise_stationary

property MarkovSwitchingARResult.is_regimewise_stationary: bool

Whether every regime is stationary read as a linear autoregression.

Sufficient for stationarity of the switching process but not necessary. A Markov-switching model with one explosive regime is stationary provided that regime is visited rarely enough and exited fast enough – the condition involves the transition matrix and the explosive root jointly, not the roots alone. A False here is a prompt to look at that regime’s expected duration, not a verdict that the process explodes.

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)
>>> res.is_regimewise_stationary
True