SETARResult.simulate

SETARResult.simulate(n=200, *, seed=None, burn=_SIMULATION_BURN)

A fresh sample path at the fitted parameters.

The regime at each step is read from the simulated series’ own delay-th lag with the estimator’s z <= r assignment, so only a self-exciting fit can simulate; a TAR fit’s transition variable is an external series whose future is not part of the model. Innovations are Gaussian at the fitted sigma2.

Parameters:

• n (int) – Observations kept.

• seed (int | Generator | None) – Seed or generator.

• burn (int) – Periods discarded from the start.

Return type:

NDArray[double]

Returns:

An array of shape (n,).

Raises:

SpecificationError – If the counts are not usable or the fit is not self-exciting.

Example

>>> import numpy as np
>>> rng = np.random.default_rng(0)
>>> y = np.zeros(400)
>>> for t in range(1, 400):
...     y[t] = (0.6 if y[t - 1] <= 0.0 else -0.4) * y[t - 1] + rng.standard_normal()
>>> res = SETAR(y, order=1, delay=1).fit()
>>> path = res.simulate(300, seed=3)
>>> path.shape, bool(np.array_equal(path, res.simulate(300, seed=3)))
((300,), True)
>>> share_lower = np.mean(path <= res.threshold)
>>> bool(abs(share_lower - res.n_lower / res.nobs) < 0.1)
True
>>> z = rng.standard_normal(400)
>>> x = np.zeros(400)
>>> for t in range(1, 400):
...     x[t] = (0.5 if z[t - 1] <= 0.0 else -0.5) * x[t - 1] + rng.standard_normal()
>>> tar = TAR(x, order=1, delay=1, threshold_variable=z).fit()
>>> tar.simulate(10)
Traceback (most recent call last):
cultivars.exceptions.SpecificationError: a TAR fit cannot simulate its own sample: ...