UnobservedComponentsResult._component_std#

UnobservedComponentsResult._component_std(name)[source]#

One component’s smoothed standard deviation path.

Single-state components read the diagonal of their block; the seasonal propagates the block covariance through the design row, so the harmonics’ covariances are counted. The caller is responsible for the component existing.

Parameters:

name (str) – "level", "slope", "cycle" or "seasonal".

Return type:

NDArray[double]

Returns:

The smoothed standard deviations, shape (n,), floored at zero against rounding.

Example

>>> import numpy as np
>>> rng = np.random.default_rng(0)
>>> y = np.cumsum(0.2 * rng.standard_normal(200)) + rng.standard_normal(200)
>>> res = UnobservedComponents(y, trend="level").fit()
>>> std = res._component_std("level")
>>> std.shape, bool(np.all(std > 0)), bool(std[0] > std[100])
((200,), True, True)