arch.univariate.distribution.Distribution.loglikelihood

abstract Distribution.loglikelihood(parameters: Sequence[float] | ndarray | Series, resids: ndarray | DataFrame | Series, sigma2: ndarray | DataFrame | Series, individual: bool = False) float | ndarray[Any, dtype[float64]][source]

Loglikelihood evaluation.

Parameters:
parameters: Sequence[float] | ndarray | Series

Distribution shape parameters

resids: ndarray | DataFrame | Series

nobs array of model residuals

sigma2: ndarray | DataFrame | Series

nobs array of conditional variances

individual: bool = False

Flag indicating whether to return the vector of individual log likelihoods (True) or the sum (False)

Notes

Returns the loglikelihood where resids are the “data”, and parameters and sigma2 are inputs.