VARMAResult._trailing_blocks#

VARMAResult._trailing_blocks()[source]#

The moving-average coefficients, in design order.

Return type:

tuple[NDArray[double], ...]

Returns:

One (k, k) block per innovation lag, each transposed so its rows match the lagged-innovation columns at the end of design.

Example

>>> import numpy as np
>>> y = np.random.default_rng(0).standard_normal((200, 2))
>>> res = VARMA(y, order=(1, 2)).fit()
>>> [block.shape for block in res._trailing_blocks()]
[(2, 2), (2, 2)]
>>> bool(np.allclose(res._trailing_blocks()[1], res.ma_coefficients[1].T))
True