arch.univariate.MIDASHyperbolic.update¶

MIDASHyperbolic.update(index: int, parameters: ndarray[tuple[int], dtype[float64]], resids: ndarray[tuple[int], dtype[float64]], sigma2: ndarray[tuple[int], dtype[float64]], backcast: float | ndarray[tuple[int], dtype[float64]], var_bounds: ndarray[tuple[int, int], dtype[float64]]) → float¶

Compute the variance for a single observation.

Parameters:¶

index: int¶

The numerical index of the variance to compute.

parameters: ndarray[tuple[int], dtype[float64]]¶

The variance model parameters.

resids: ndarray[tuple[int], dtype[float64]]¶

The residual array. Only uses resids[:index] when computing sigma2[index].

sigma2: ndarray[tuple[int], dtype[float64]]¶

The array containing the variances. Only uses sigma2[:index] when computing sigma2[index]. The computed value is stored in sigma2[index].

backcast: float | ndarray[tuple[int], dtype[float64]]¶

Value to use when initializing the recursion.

var_bounds: ndarray[tuple[int, int], dtype[float64]]¶

Array containing columns of lower and upper bounds.

Returns:¶

The variance computed for location index.

Return type:¶

float