FitControlsBicop.__init__

FitControlsBicop.__init__(self, family_set: collections.abc.Sequence[pyvinecopulib.families.BicopFamily] = [BicopFamily.indep, BicopFamily.gaussian, BicopFamily.student, BicopFamily.clayton, BicopFamily.gumbel, BicopFamily.frank, BicopFamily.joe, BicopFamily.bb1, BicopFamily.bb6, BicopFamily.bb7, BicopFamily.bb8, BicopFamily.tawn, BicopFamily.tll], parametric_method: str = 'mle', nonparametric_method: str = 'constant', nonparametric_mult: float = 1.0, nonparametric_grid_size: int = 30, selection_criterion: str = 'bic', weights: numpy.ndarray = array([], dtype=float64), psi0: float = 0.9, preselect_families: bool = True, allow_rotations: bool = True, num_threads: int = 1) None

Instantiates the controls for fitting bivariate copula models.

Parameters:
family_setlist of BicopFamily

The set of copula families to consider (if empty, then all families are included).

parametric_methodstr

The fit method for parametric families; possible choices: "mle", "itau".

nonparametric_methodstr

The fit method for the local-likelihood nonparametric family (TLLs); possible choices: "constant", "linear", "quadratic".

nonparametric_multfloat

A factor with which the smoothing parameters are multiplied (default: 1.0).

nonparametric_grid_sizeint

The grid size for the post-estimation interpolation in nonparametric models (default: 30).

selection_criterionstr

The selection criterion ("loglik", "aic", "bic", "mbic", or "mbicv") for the pair copula families.

weightsndarray, shape (n,), dtype float

A vector of weights for the observations.

psi0float

Only for selection_criterion = "mbic", the prior probability of non-independence.

preselect_familiesbool

Whether to exclude families before fitting based on symmetry properties of the data.

allow_rotationsbool

Allow rotations for the families when doing model selection (default: true).

num_threadsint

Number of concurrent threads to use while fitting copulas for different families; never uses more than the number of concurrent threads supported by the implementation.