ThetaModularForecaster

class ThetaModularForecaster(forecasters=None, theta_values=(0, 2), aggfunc='mean', weights=None)[source]

Modular theta method for forecasting.

Modularized implementation of Theta method as defined in [1]. Also see auto-theta method as described in [2] *not contained in this estimator).

Overview: Input univariate series of length “n” and decompose with ThetaLinesTransformer by modifying the local curvature of the time series using Theta-coefficient values - theta_values parameter. Thansformation gives a pd.DataFrame of shape len(input series) * len(theta).

The resulting transformed series (Theta-lines) are extrapolated separately. The forecasts are then aggregated into one prediction - aunivariate series, of len(fh).

Parameters:

forecasters: list of tuples (str, estimator, int or pd.index), default=None: Forecasters to apply to each Theta-line based on the third element (the index). Indices must correspond to the theta_values, see Examples. If None, will apply PolynomialTrendForecaster (linear regression) to the Theta-lines where theta_value equals 0, and ExponentialSmoothing - where theta_value is different from 0.
theta_values: sequence of float, default=(0,2): Theta-coefficients to use in transformation. If forecasters parameter is passed, must be the same length as forecasters.
aggfunc: str, default=”mean”: Must be one of [“mean”, “median”, “min”, “max”, “gmean”]. Calls _aggregate of EnsembleForecaster to apply to results of multivariate theta-lines predictions (pd.DataFrame) in order to get resulting univariate prediction (pd.Series). The aggregation takes place across different theta-lines (row-wise), for given time stamps and hierarchy indices, if present.
weights: list of floats, default=None: Weights to apply in aggregation. Weights are passed as a parameter to the aggregation function, must correspond to each theta-line. None will result in non-weighted aggregation.

Attributes:

cutoff: Cut-off = “present time” state of forecaster.
fh: Forecasting horizon that was passed.
is_fitted: Whether fit has been called.
state: State of the estimator.

References

[1]

V.Assimakopoulos et al., “The theta model: a decomposition approach to forecasting”, International Journal of Forecasting, vol. 16, pp. 521-530, 2000.

[2]

E.Spiliotis et al., “Generalizing the Theta method for automatic forecasting “, European Journal of Operational Research, vol. 284, pp. 550-558, 2020.

Examples

>>> from sktime.datasets import load_airline
>>> from sktime.forecasting.theta import ThetaModularForecaster
>>> from sktime.forecasting.naive import NaiveForecaster
>>> from sktime.forecasting.trend import PolynomialTrendForecaster
>>> y = load_airline()
>>> forecaster = ThetaModularForecaster(
...     forecasters=[
...         ("trend", PolynomialTrendForecaster(), 0),
...         ("arima", NaiveForecaster(), 3),
...     ],
...     theta_values=(0, 3),
... )
>>> forecaster.fit(y)
ThetaModularForecaster(...)
>>> y_pred = forecaster.predict(fh=[1,2,3])

Methods

`check_is_fitted`([method_name])	Check if the estimator has been fitted.
`clone`()	Obtain a clone of the object with same hyper-parameters and config.
`clone_tags`(estimator[, tag_names])	Clone tags from another object as dynamic override.
`create_test_instance`([parameter_set])	Construct an instance of the class, using first test parameter set.
`create_test_instances_and_names`([parameter_set])	Create list of all test instances and a list of names for them.
`fit`(y[, X, fh])	Fit forecaster to training data.
`fit_predict`(y[, X, fh, X_pred])	Fit and forecast time series at future horizon.
`get_class_tag`(tag_name[, tag_value_default])	Get class tag value from class, with tag level inheritance from parents.
`get_class_tags`()	Get class tags from class, with tag level inheritance from parent classes.
`get_config`()	Get config flags for self.
`get_fitted_params`([deep])	Get fitted parameters.
`get_param_defaults`()	Get object's parameter defaults.
`get_param_names`([sort])	Get object's parameter names.
`get_params`([deep])	Get a dict of parameters values for this object.
`get_pretrained_params`([deep])	Get pretrained parameters of this estimator.
`get_tag`(tag_name[, tag_value_default, ...])	Get tag value from instance, with tag level inheritance and overrides.
`get_tags`()	Get tags from instance, with tag level inheritance and overrides.
`get_test_params`([parameter_set])	Return testing parameter settings for the estimator.
`is_composite`()	Check if the object is composed of other BaseObjects.
`load_from_path`(serial)	Load object from file location.
`load_from_serial`(serial)	Load object from serialized memory container.
`predict`([fh, X])	Forecast time series at future horizon.
`predict_interval`([fh, X, coverage])	Compute/return prediction interval forecasts.
`predict_proba`([fh, X, marginal])	Compute/return fully probabilistic forecasts.
`predict_quantiles`([fh, X, alpha])	Compute/return quantile forecasts.
`predict_residuals`([y, X])	Return residuals of time series forecasts.
`predict_var`([fh, X, cov])	Compute/return variance forecasts.
`pretrain`(y[, X, fh])	Pre-train forecaster on panel (global) data.
`reset`()	Reset the object to a clean post-init state.
`save`([path, serialization_format])	Save serialized self to bytes-like object or to (.zip) file.
`score`(y[, X, fh])	Scores forecast against ground truth, using MAPE (non-symmetric).
`set_config`(**config_dict)	Set config flags to given values.
`set_params`(**params)	Set the parameters of this object.
`set_random_state`([random_state, deep, ...])	Set random_state pseudo-random seed parameters for self.
`set_tags`(**tag_dict)	Set instance level tag overrides to given values.
`update`(y[, X, update_params])	Update cutoff value and, optionally, fitted parameters.
`update_predict`(y[, cv, X, update_params, ...])	Make predictions and update model iteratively over the test set.
`update_predict_single`([y, fh, X, update_params])	Update model with new data and make forecasts.