MIDASVAR._seeds

MIDASVAR._seeds()

Starting points: flat weights, gentle decay, and steep decay.

Return type:

tuple[NDArray[double], ...]

Returns:

Three (2 * m,) vectors with theta_1 = 0 for every series and theta_2 in (0.0, -0.05, -0.5).

Example

>>> import numpy as np
>>> rng = np.random.default_rng(0)
>>> x, y = rng.standard_normal((60, 2)), rng.standard_normal((20, 2))
>>> [seed.tolist() for seed in MIDASVAR(y, x, order=1, period=3)._seeds()]
[[0.0, 0.0, 0.0, 0.0], [0.0, -0.05, 0.0, -0.05], [0.0, -0.5, 0.0, -0.5]]