Back to models
Forecaster

StatsForecastAutoTBATS

API referenceSource
Categorical in XInsamplePred intPred int insample

StatsForecast TBATS model.

Direct interface to statsforecast.models.AutoTBATS, from statsforecast [1] by Nixtla.

Automatically selects the best TBATS model from all feasible combinations of the parameters use_boxcox, use_trend, use_damped_trend, and use_arma_errors. Selection is made using the AIC.

Default value for use_arma_errors is True since this enables the evaluation of models with and without ARMA errors.

Quickstart

python
from sktime.forecasting.statsforecast import StatsForecastAutoTBATS

estimator = StatsForecastAutoTBATS(seasonal_periods: int | list [int ]=1, use_boxcox: bool | None=None, use_trend: bool | None=None, use_damped_trend: bool | None=None, use_arma_errors: bool=True, bc_lower_bound: float=0.0, bc_upper_bound: float=1.0)

Parameters(7)

seasonal_periodsint or list of int. (default=1)
Number of observations per unit of time. Ex: 24 Hourly data.
use_boxcoxbool (default=None)
Whether or not to use a Box-Cox transformation. By default tries both.
use_trendbool (default=None)
Whether or not to use a trend component. By default tries both.
use_damped_trendbool (default=None)
Whether or not to dampen the trend component. By default tries both.
use_arma_errorsbool (default=True)
Whether or not to use a ARMA errors. Default is True and this evaluates both models.
bc_lower_boundfloat (default=0.0)
Lower bound for the Box-Cox transformation.
bc_upper_boundfloat (default=1.0)
Upper bound for the Box-Cox transformation.

References