larch.ModelGroup.d_loglike

ModelGroup.d_loglike(x=None, *, return_series=False, **kwargs)

Compute the gradient of the log likelihood of the group of models.

Note that unlike for a single model, the gradient of the log likelihood cannot be computed for a slice of cases, but must be computed for all cases.

Parameters:

x (array-like or dict, optional) – New values to set for the parameters before evaluating the gradient of the log likelihood. If given as array-like, the array must be a vector with length equal to the length of the parameter frame, and the given vector will replace the current values. If given as a dictionary, the dictionary is used to update the parameters.

Returns:

array-like