bilby.core.likelihood.Multinomial

class bilby.core.likelihood.Multinomial(data, n_dimensions, base='parameter_')

Bases: Likelihood

Likelihood for system with N discrete possibilities.

__init__(data, n_dimensions, base='parameter_')

Parameters:

data: array-like

The number of objects in each class

n_dimensions: int

The number of classes

base: str

The base of the parameter labels

__call__(*args, **kwargs)

Call self as a function.

Methods

__init__(data, n_dimensions[, base])
log_likelihood(parameters)	Since n - 1 parameters are sampled, the last parameter is 1 - the rest
log_likelihood_ratio(parameters)	Difference between log likelihood and noise log likelihood
noise_log_likelihood()	Our null hypothesis is that all bins have probability 1 / nbins, i.e., no bin is preferred over any other.

Attributes

marginalized_parameters
meta_data

log_likelihood(parameters)

Since n - 1 parameters are sampled, the last parameter is 1 - the rest

Parameters:

parameters: dict

A dictionary of the parameter names and associated values

Returns:

float

log_likelihood_ratio(parameters)

Difference between log likelihood and noise log likelihood

Parameters:

parameters: dict

A dictionary of the parameter names and associated values

Returns:

float

noise_log_likelihood()

Our null hypothesis is that all bins have probability 1 / nbins, i.e., no bin is preferred over any other.