statsmodels.distributions.discrete.DiscretizedModel.predict

DiscretizedModel.predict(params, exog=None, which=None, k_max=20)

Predict values for the discretized distribution.

Parameters:

paramsarray_like

Shape (and scale) parameters of the underlying continuous distribution.

exogNone

Explanatory variables are not supported. The exog argument is only included for consistency in the signature across models.

whichstr, optional

Statistic to predict. The only currently implemented value is “probs”, which returns the probability mass function evaluated at 0, 1, ..., k_max - 1. Any other value, including the default of None, raises a ValueError.

k_maxint, optional

Number of support points, starting at zero, for which probabilities are returned.

Returns:

ndarray

Probability mass function evaluated at 0, 1, ..., k_max - 1.