Censored#

class pymc_marketing.prior.Censored(*args, **kwargs)[source]#

Create censored random variable.

Examples

Create a censored Normal distribution:

from pymc_marketing.prior import Prior, Censored

normal = Prior("Normal")
censored_normal = Censored(normal, lower=0)

Create hierarchical censored Normal distribution:

from pymc_marketing.prior import Prior, Censored

normal = Prior(
    "Normal",
    mu=Prior("Normal"),
    sigma=Prior("HalfNormal"),
    dims="channel",
)
censored_normal = Censored(normal, lower=0)

coords = {"channel": range(3)}
samples = censored_normal.sample_prior(coords=coords)

Methods

`Censored.__init__`(args, *kwargs)
`Censored.create_likelihood_variable`(name, ...)	Create observed censored variable.
`Censored.create_variable`(name)	Create censored random variable.
`Censored.from_dict`(data)	Create a censored distribution from a dictionary.
`Censored.sample_prior`([coords, name])	Sample the prior distribution for the variable.
`Censored.to_dict`()	Convert the censored distribution to a dictionary.
`Censored.to_graph`()	Generate a graph of the variables.

Attributes

`dims`	The dims from the distribution to censor.
`lower`
`upper`
`distribution`