Back to models
Forecaster

IgnoreX

API referenceSource
Categorical featuresIn-sample predictionsIn-sample prediction intervalsExogenous variables

Compositor for ignoring exogenous variables.

Composing with IgnoreX instructs the wrapped forecaster to ignore exogenous data. This is useful for testing the impact of exogenous data on forecasts, or for use in tuning hyperparameters of the forecaster.

Quickstart

python
from sktime.forecasting.compose import IgnoreX

estimator = IgnoreX(forecaster, ignore_x=True)

Tags

Capabilities

  • Categorical features
  • In-sample predictions: Supported
  • In-sample prediction intervals: Supported
  • Exogenous variables
  • Prediction intervals: Not supported
  • Missing values: Not supported
  • Multivariate: Not supported

Properties

Needs forecast horizon in fitrequires-fh-in-fit
Yes
X and y need the same indexX-y-must-have-same-index
Yes

Parameters(2)

forecastersktime forecaster, BaseForecaster descendant instance
The forecaster to wrap.
ignore_xbool, optional (default=True)

Whether to ignore exogenous data or not, this parameter is useful for tuning.

  • True: ignore exogenous data, X is not passed on to forecaster

  • False: use exogenous data, X is passed on to forecaster

Examples

>>> from sktime.forecasting.compose import IgnoreX
>>> from sktime.forecasting.sarimax import SARIMAX
>>> from sktime.datasets import load_longley
>>> y, X = load_longley ()
>>> forecaster = IgnoreX (SARIMAX ())
>>> forecaster. fit (y, X = X, fh = [1, 2, 3 ]) IgnoreX(forecaster=SARIMAX())
>>> y_pred = forecaster. predict (X = X)