AutoARIMAModel

class AutoARIMAModel(**kwargs)

Bases: etna.models.mixins.PerSegmentModelMixin, etna.models.mixins.PredictionIntervalContextIgnorantModelMixin, etna.models.base.PredictionIntervalContextIgnorantAbstractModel

Class for holding auto arima model.

Method predict can use true target values only on train data on future data autoregression forecasting will be made even if targets are known.

Notes

We use pmdarima.arima.arima.ARIMA.

This model supports in-sample and out-of-sample prediction decomposition. Prediction components for AutoARIMA model are: exogenous and ARIMA components. Decomposition is obtained directly from fitted model parameters.

Init auto ARIMA model with given params.

Parameters

**kwargs – Training parameters for auto_arima from pmdarima package.

Inherited-members

Methods

fit(ts)	Fit model.
forecast(ts[, prediction_interval, ...])	Make predictions.
get_model()	Get internal models that are used inside etna class.
load(path)	Load an object.
params_to_tune()	Get grid for tuning hyperparameters.
predict(ts[, prediction_interval, ...])	Make predictions with using true values as autoregression context if possible (teacher forcing).
save(path)	Save the object.
set_params(**params)	Return new object instance with modified parameters.
to_dict()	Collect all information about etna object in dict.

Attributes

context_size

Context size of the model.