redback.likelihoods.GaussianLikelihoodWithSystematicNoise

class redback.likelihoods.GaussianLikelihoodWithSystematicNoise(*args: Any, **kwargs: Any)[source]

Bases: GaussianLikelihood

__init__(x: ndarray, y: ndarray, sigma_i: float | None | ndarray, function: callable, kwargs: dict = None, priors=None, fiducial_parameters=None) → None[source]

A Gaussian likelihood with a systematic noise term that is proportional to the model + the original data noise added in quadrature. The parameters are inferred from the arguments of function

Parameters:

  • x (np.ndarray) – The x values.

  • y (np.ndarray) – The y values.

  • sigma_i (Union[float, None, np.ndarray]) – The standard deviation of the noise. This is part of the full noise. The sigma used in the likelihood is sigma = sqrt(sigma_i^2 + model_y**2*sigma^2)

  • function (callable) – The python function to fit to the data. Note, this must take the dependent variable as its first argument. The other arguments will require a prior and will be sampled over (unless a fixed value is given).

  • kwargs (dict) – Any additional keywords for ‘function’.

  • priors – The priors for the parameters. Default to None if not provided.

Only necessary if using maximum likelihood estimation functionality. :type priors: Union[dict, None] :param fiducial_parameters: The starting guesses for model parameters to use in the optimization for maximum likelihood estimation. Default to None if not provided. :type fiducial_parameters: Union[dict, None]

__call__(*args: Any, **kwargs: Any) → Any

Call self as a function.

Methods

__init__(x, y, sigma_i, function[, kwargs, ...])

A Gaussian likelihood with a systematic noise term that is proportional to the model + the original data noise added in quadrature.

find_maximum_likelihood_parameters([...])

Estimate the maximum likelihood

get_bounds_from_priors(priors)

get_parameter_dictionary_from_list(...)

get_parameter_list_from_dictionary(...)

lnlike_scipy_maximize(parameter_list)

log_likelihood([parameters])

noise_log_likelihood()

Attributes

full_sigma

The standard deviation of the full noise :rtype: Union[float, np.ndarray]

kwargs

model_output

The model output for the given x values.

n

Length of the x/y-values :rtype: int

parameters

parameters_to_be_updated

residual

sigma

find_maximum_likelihood_parameters(iterations=5, maximization_kwargs=None, method='Nelder-Mead', break_threshold=0.001)

Estimate the maximum likelihood

Parameters:

  • iterations – Iterations to run the minimizer for before stopping. Default is 5.

  • maximization_kwargs – Any extra keyword arguments passed to the scipy minimize function

  • method – Minimize method to use. Default is ‘Nelder-Mead’

  • break_threshold – The threshold for the difference in log likelihood to break the loop. Default is 1e-3.

Returns:

Dictionary of maximum likelihood parameters

property full_sigma: float | ndarray

The standard deviation of the full noise :rtype: Union[float, np.ndarray]

Type:

return

log_likelihood(parameters=None) → float[source]
Returns:

The log-likelihood.

Return type:

float

property model_output: ndarray

The model output for the given x values. :rtype: np.ndarray

Type:

return

property n: int

Length of the x/y-values :rtype: int

Type:

return

noise_log_likelihood() → float[source]
Returns:

The noise log-likelihood, i.e. the log-likelihood assuming the signal is just noise.

Return type:

float