MFVAR._state_space

MFVAR._state_space()

Assemble the observable form.

Builds the stacked system through _mixed_frequency_system – a companion transition over max(order, period) lags, a selection onto the current block, and a constant design matrix carrying each variable’s aggregation weights – and wraps it in the package’s linear Gaussian state space.

Return type:

_LinearGaussianStateSpace

Returns:

The state-space model whose filter and smooth the public methods call.

Example

>>> import numpy as np
>>> A, S = np.array([[[0.5, 0.1], [0.0, 0.4]]]), np.eye(2)
>>> model = MFVAR(A, S, kinds=["high", "flow"], period=3)
>>> system = model._state_space()
>>> type(system).__name__, system.filter(np.zeros((4, 2))).filtered_state.shape
('_LinearGaussianStateSpace', (4, 6))