Back to models
Forecaster

StatsForecastAutoTheta

API referenceSource
In-sample predictionsPrediction intervalsIn-sample prediction intervals

Statsforecast AutoTheta estimator.

Direct interface to statsforecast.models.AutoTheta by Nixtla.

This estimator directly interfaces AutoTheta, from statsforecast [1] by Nixtla.

AutoTheta model automatically selects the best Theta (Standard Theta Model (“STM”), Optimized Theta Model (“OTM”), Dynamic Standard Theta Model (“DSTM”), Dynamic Optimized Theta Model (“DOTM”)) model using mse.

Quickstart

python
from sktime.forecasting.statsforecast import StatsForecastAutoTheta

estimator = StatsForecastAutoTheta(season_length: int=1, decomposition_type: str='multiplicative', model: str | None=None)

Tags

Capabilities

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

Properties

Needs forecast horizon in fitrequires-fh-in-fit
No
X and y need the same indexX-y-must-have-same-index
Yes

Parameters(3)

season_lengthint, optional, default=1
number of observations per unit of time (e.g. 24 for hourly data), by default 1
decomposition_typestr, optional, default=”multiplicative”
possible values: “additive”, “multiplicative” type of seasonal decomposition, by default “multiplicative”
modelOptional[str], optional
controlling Theta Model, by default searches the best model possible values: “STM”, “OTM”, “DSTM”, “DOTM”

Examples

>>> from sktime.datasets import load_airline
>>> from sktime.forecasting.statsforecast import StatsForecastAutoTheta
>>> y = load_airline ()
>>> forecaster = StatsForecastAutoTheta (
... season_length = 12, decomposition_type = "additive", model = "OTM"
... )
>>> forecaster. fit (y) StatsForecastAutoTheta(
... )
>>> y_pred = forecaster. predict (fh = [1, 2, 3 ])

References