models.pit()

Probability integral transform of each outcome under the family’s

Usage

models.pit(
    family,
    y,
    mu,
    dispersion=1.0,
    weights=None,
    seed=0,
    theta=None,
    power=None
)

predictive distribution, randomized where it has atoms (counts, a Tweedie’s zero); uniform when the model is calibrated.

Parameters

family: str

y: list of float

mu: list of float

dispersion: float = 1.0

weights: list of float = None

seed: int = 0

Seeds the randomization.

theta: float = None

power: float = None

Returns

list of float