Likelihoods for StatsForecastModel#

class darts.utils.likelihood_models.statsforecast.QuantilePrediction(quantiles)[source]#

Bases: Likelihood, ABC

Quantile Prediction Likelihood

Can be used to generate quantile predictions for any Darts model that wraps a statsforecast model.

Parameters:: quantiles (list[float]) – A list of quantiles. Default: [0.01, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 0.99].

Attributes

`num_parameters`	Returns the number of distribution parameters for a single target value.
`parameter_names`	Returns the likelihood parameter names.
`type`	Returns the likelihood type.

Methods

`component_names`([series, components])	Generates names for the parameters of the Likelihood.
`fit`(model)	Fits the likelihood to the model.
`predict`(model_output, num_samples)	Generates sampled or direct likelihood parameter predictions.
`sample`(model_output, num_samples)	Samples a prediction from the likelihood distribution and the predicted parameters.

component_names(series=None, components=None)#

Generates names for the parameters of the Likelihood.

property num_parameters: int#: Returns the number of distribution parameters for a single target value.

property parameter_names: list[str]#: Returns the likelihood parameter names.

predict(model_output, num_samples)[source]#

Generates sampled or direct likelihood parameter predictions.

Parameters:

model_output (ndarray) – The output of the StatsForecast model.
num_samples (int) – Number of times a prediction is sampled from the likelihood model / distribution. If 1 and predict_likelihood_parameters=False, returns median / mean predictions.

Return type:

ndarray

sample(model_output, num_samples)[source]#

Samples a prediction from the likelihood distribution and the predicted parameters.