SumOfCoefficientsPrior.coefficient_variance#

SumOfCoefficientsPrior.coefficient_variance(context)[source]#

Infinite: no diagonal opinion about any coefficient.

Infinity is how a prior says it has no view of a coefficient, and in composition an infinite variance contributes zero precision, so stacking this component onto a proper prior leaves that prior’s moments exactly as they were. It is also why the component cannot stand alone: a marginal likelihood needs every variance finite.

Parameters:

context (_PriorContext) – The sample description; width and k_endog are read.

Return type:

NDArray[double]

Returns:

A (width, k) array of inf.

Example

>>> import numpy as np
>>> from cultivars.bayes.priors import PriorContext
>>> ctx = PriorContext(
...     k_endog=2, order=1, scales=np.ones(2), presample_mean=np.zeros(2)
... )
>>> bool(np.isinf(SumOfCoefficientsPrior().coefficient_variance(ctx)).all())
True