TVPVARResult._drift_note#

TVPVARResult._drift_note()[source]#

One line on how much the coefficients actually moved.

Return type:

str

Returns:

The largest absolute change of any posterior mean coefficient between the first and last date, set against the median width of the 68% bands, with the reminder that drift inside the band is not evidence of time variation.

Example

>>> import numpy as np
>>> y = np.random.default_rng(0).standard_normal((120, 2))
>>> res = TVPVAR(y, order=1, training=30).fit(n_draws=20, n_burn=8, seed=0)
>>> res._drift_note().startswith("Posterior mean drift over the sample")
True