Vinecop.bic
- Vinecop.bic(self, u: numpy.ndarray[dtype=float64, shape=(*, *), order='F'] = array([], shape=(0, 0), dtype=float64), num_threads: int = 1) float
Evaluates the Bayesian information criterion (BIC).
The BIC is defined as
where
is the log-liklihood (seeVinecop.loglik()
) and is the (effective) number of parameters of the model. The BIC is a consistent model selection criterion for nonparametric models.- Parameters:
u – An
or matrix of evaluation points, where is the number of discrete variables (seeVinecop.select()
orVinecop.pdf()
).num_threads – The number of threads to use for computations; if greater than 1, the function will be applied concurrently to
num_threads
batches ofu
.
- Returns:
The BIC as a double.