Skip to content

Priors

Prior specifications for VAR models.

MinnesotaPrior

Bases: BaseModel

Minnesota prior for VAR coefficient shrinkage.

Attributes:

Name Type Description
tightness float

Overall shrinkage toward prior mean. Must be > 0.

decay Literal['harmonic', 'geometric']

How coefficients shrink on longer lags.

cross_shrinkage float

Shrinkage on other variables' lags vs own. 0 = only own lags, 1 = equal.

build_priors(n_vars, n_lags)

Build prior mean and standard deviation arrays for VAR coefficients.

Parameters:

Name Type Description Default
n_vars int

Number of endogenous variables.

required
n_lags int

Number of lags.

required

Returns:

Type Description
dict

Dictionary with keys 'B_mu' and 'B_sigma' as numpy arrays.