arch.univariate.distribution.Distribution.loglikelihood

abstract Distribution.loglikelihood(parameters, resids, sigma2, individual=False)[source]

Loglikelihood evaluation.

Parameters

  • parameters (ndarray) -- Distribution shape parameters

  • resids (ndarray) -- nobs array of model residuals

  • sigma2 (ndarray) -- nobs array of conditional variances

  • individual (bool, optional) -- 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.

Return type

Union[float, ndarray]