DynamicNelsonSiegelResult.curve#

DynamicNelsonSiegelResult.curve(date_index, maturities=None)[source]#

The fitted curve at one date, on any maturity grid.

Parameters:

  • date_index (int) – Row of the panel (negative indexing allowed).

  • maturities (ArrayLike | None) – Strictly positive maturities to evaluate on; defaults to the estimation grid.

Return type:

NDArray[double]

Returns:

Fitted yields at those maturities.

Raises:

SpecificationError – If a requested maturity is not strictly positive.

Example

A pure-level date prices every maturity at the level; a negative slope makes the curve rise with maturity:

>>> import numpy as np
>>> paths = np.tile([5.0, -1.5, 0.0], (30, 1))
>>> 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.eye(3), measurement_var=np.full(4, 0.01),
...     factors=paths, factor_cov=np.tile(np.eye(3), (30, 1, 1)),
...     llf=0.0, nobs=30, n_params=16.0,
... )
>>> fine = res.curve(-1, maturities=np.linspace(0.25, 30.0, 50))
>>> fine.shape, bool(np.all(np.diff(fine) > 0)), bool(fine[-1] < 5.0)
((50,), True, True)
>>> res.curve(0, maturities=[0.0, 1.0])
Traceback (most recent call last):
cultivars.exceptions.SpecificationError: maturities must be strictly positive.