PanelVARResult.unit_series#

PanelVARResult.unit_series(unit)[source]#

One unit’s observations, including the rows consumed as lags.

Parameters:

unit (str) – A unit label.

Return type:

NDArray[double]

Returns:

A (T_i, k) view into endog.

Raises:

SpecificationError – If the unit is unknown.

Example

>>> import numpy as np
>>> rng = np.random.default_rng(0)
>>> units = [rng.standard_normal((n, 2)) for n in (60, 45)]
>>> res = PanelVAR(units, order=1).fit()
>>> series = res.unit_series("unit2")
>>> series.shape, bool(np.allclose(series, units[1]))
((45, 2), True)
>>> bool(np.shares_memory(series, res.endog))
True