GVARResult.forecast

GVARResult.forecast(steps=1)

Deterministic multi-step forecasts from the end of the sample.

Iterates the estimated system forward with future innovations set to their zero mean, which is the conditional expectation. No interval is returned: the coefficient uncertainty that should widen it is exactly the standard-error machinery this package does not yet have, and a band computed as though the coefficients were known would be too narrow in a way nobody could see.

Parameters:

steps (int) – Forecast horizon, at least one.

Return type:

NDArray[double]

Returns:

An array of shape (steps, k).

Raises:

SpecificationError – If steps is less than one.