DynamicNelsonSiegelResult._params#

DynamicNelsonSiegelResult._params[source]#

The parameter record, rebuilt for the system builder.

Returns:

A _NelsonSiegelParameters with the Cholesky factor of state_innovation_cov in place of the covariance, which is the form the state-space builder and the objective share.

Example

>>> import numpy as np
>>> res = DynamicNelsonSiegelResult(
...     panel=np.zeros((30, 4)), maturities=np.array([1.0, 2.0, 5.0, 10.0]),
...     decay=0.6, decay_fixed=True, mu=np.zeros(3), ar=np.full(3, 0.9),
...     state_innovation_cov=np.diag([0.04, 0.09, 0.16]),
...     measurement_var=np.full(4, 0.01), factors=np.zeros((30, 3)),
...     factor_cov=np.tile(np.eye(3), (30, 1, 1)), llf=0.0, nobs=30, n_params=16.0,
... )
>>> params = res._params
>>> params.decay, np.diag(params.state_chol).round(2)
(0.6, array([0.2, 0.3, 0.4]))