MarkovSwitchingVARResult._likelihood_ratio_obstacle

MarkovSwitchingVARResult._likelihood_ratio_obstacle(counterpart)

Block only the comparison that changes the number of regimes.

A test holding M fixed – dropping a switching block, say – is perfectly valid; testing M itself is not, because under the smaller model the extra regime’s parameters are unidentified and its transition probabilities sit on the boundary. A non-switching result counts as the one-regime case, which is exactly the classic invalid test.

Parameters:

counterpart (_ComparisonMixin) – The other result in the comparison.

Return type:

str | None

Returns:

None when the test may proceed, else the refusal message.

Example

>>> import numpy as np
>>> from cultivars.multivariate.reduced_form.vector_autoregression import VAR
>>> rng = np.random.default_rng(0)
>>> s = np.repeat([0, 1, 0, 1], 100)
>>> y = np.array([[-1.0, -0.5], [1.5, 1.0]])[s] + 0.4 * rng.standard_normal((400, 2))
>>> full = MarkovSwitchingVAR(y, order=1, n_regimes=2).fit(seed=0)
>>> model = MarkovSwitchingVAR(y, order=1, n_regimes=2, switching_variance=False)
>>> same_m = model.fit(seed=0)
>>> full._likelihood_ratio_obstacle(same_m) is None
True
>>> full._likelihood_ratio_obstacle(VAR(y, order=1).fit())[:60]
'a chi-squared likelihood-ratio test cannot compare 2 regimes'
>>> same_m.likelihood_ratio_test(full).df
3