Forecaster

ThetaForecaster

Categorical in XInsamplePred intPred int insample

Theta method for forecasting.

The theta method as defined in [1] is equivalent to simple exponential smoothing (SES) with drift (as demonstrated in [2]).

The series is tested for seasonality using the test outlined in A&N. If deemed seasonal, the series is seasonally adjusted using a classical multiplicative decomposition before applying the theta method. The resulting forecasts are then reseasonalised.

In cases where SES results in a constant forecast, the theta forecaster will revert to predicting the SES constant plus a linear trend derived from the training data.

Prediction intervals are computed using the underlying state space model.

Schnellstart

python

from sktime.forecasting.theta import ThetaForecaster

estimator = ThetaForecaster(initial_level=None, deseasonalize=True, sp=1, deseasonalize_model='multiplicative')

Parameter(4)

initial_levelfloat, optional: The alpha value of the simple exponential smoothing, if the value is set then this will be used, otherwise it will be estimated from the data.
deseasonalizebool, optional (default=True), or sktime BaseTransformer instance: Whether and how to deseasonalize the data before fitting the theta model.
If True, data is seasonally adjusted using sktime Deseasonalizer().
If BaseTransformer instance, this is used to seasonally adjust the data, via fit_transform in fit, and inverse_transform in predict.
If False, no seasonal adjustment is done.
spint, optional (default=1): The number of observations that constitute a seasonal period for a multiplicative deseasonaliser, which is used if seasonality is detected in the training data. Ignored if a deseasonaliser transformer is provided. Default is 1 (no seasonality).
deseasonalize_modelstr, optional (default=”multiplicative”): The type of seasonal decomposition to use in the deseasonaliser. Can be “additive” or “multiplicative”. Passed on to Deseasonalizer if deseasonalize=True. Only used if deseasonalize=True.

Beispiele

>>> from sktime.datasets import load_airline
>>> from sktime.forecasting.theta import ThetaForecaster
>>> y = load_airline ()
>>> forecaster = ThetaForecaster (sp = 12)
>>> forecaster. fit (y) ThetaForecaster(
... )
>>> y_pred = forecaster. predict (fh = [1, 2, 3 ])

Referenzen

[1]

Assimakopoulos, V. and Nikolopoulos, K. The theta model: a decomposition approach to forecasting. International Journal of Forecasting 16, 521-530, 2000. https://www.sciencedirect.com/science/article/pii/S0169207000000662

[2]

`Hyndman, Rob J., and Billah, Baki. Unmasking the Theta method. International J. Forecasting, 19, 287-290, 2003. https://www.sciencedirect.com/science/article/pii/S0169207001001431