pyFTS.models.ensemble package

Submodules

pyFTS.models.ensemble.ensemble module

EnsembleFTS wraps several FTS methods to ensemble their forecasts, providing point, interval and probabilistic forecasting.

Silva, P. C. L et al. Probabilistic Forecasting with Seasonal Ensemble Fuzzy Time-Series XIII Brazilian Congress on Computational Intelligence, 2017. Rio de Janeiro, Brazil.

class pyFTS.models.ensemble.ensemble.AllMethodEnsembleFTS(**kwargs)[source]

Bases: pyFTS.models.ensemble.ensemble.EnsembleFTS

Creates an EnsembleFTS with all point forecast methods, sharing the same partitioner

set_transformations(model)[source]
train(data, **kwargs)[source]

Method specific parameter fitting

Parameters
  • data – training time series data

  • kwargs – Method specific parameters

class pyFTS.models.ensemble.ensemble.EnsembleFTS(**kwargs)[source]

Bases: pyFTS.common.fts.FTS

Ensemble FTS

alpha

The quantiles

append_model(model)[source]

Append a new trained model to the ensemble

Parameters

model – FTS model

forecast(data, **kwargs)[source]

Point forecast one step ahead

Parameters
  • data – time series data with the minimal length equal to the max_lag of the model

  • kwargs – model specific parameters

Returns

a list with the forecasted values

forecast_ahead_distribution(data, steps, **kwargs)[source]

Probabilistic forecast from 1 to H steps ahead, where H is given by the steps parameter

Parameters
  • data – time series data with the minimal length equal to the max_lag of the model

  • steps – the number of steps ahead to forecast

  • start_at – in the multi step forecasting, the index of the data where to start forecasting (default: 0)

Returns

a list with the forecasted Probability Distributions

forecast_ahead_interval(data, steps, **kwargs)[source]

Interval forecast from 1 to H steps ahead, where H is given by the steps parameter

Parameters
  • data – time series data with the minimal length equal to the max_lag of the model

  • steps – the number of steps ahead to forecast

  • start_at – in the multi step forecasting, the index of the data where to start forecasting (default: 0)

Returns

a list with the forecasted intervals

forecast_distribution(data, **kwargs)[source]

Probabilistic forecast one step ahead

Parameters
  • data – time series data with the minimal length equal to the max_lag of the model

  • kwargs – model specific parameters

Returns

a list with probabilistic.ProbabilityDistribution objects representing the forecasted Probability Distributions

forecast_interval(data, **kwargs)[source]

Interval forecast one step ahead

Parameters
  • data – time series data with the minimal length equal to the max_lag of the model

  • kwargs – model specific parameters

Returns

a list with the prediction intervals

get_UoD()[source]

Returns the interval of the known bounds of the universe of discourse (UoD), i. e., the known minimum and maximum values of the time series.

Returns

A set with the lower and the upper bounds of the UoD

get_distribution_interquantile(forecasts, alpha)[source]
get_interval(forecasts)[source]
get_models_forecasts(data)[source]
get_point(forecasts, **kwargs)[source]
interval_method

The method used to mix the several model’s forecasts into a interval forecast. Options: quantile, extremum, normal

models

A list of FTS models, the ensemble components

parameters

A list with the parameters for each component model

point_method

The method used to mix the several model’s forecasts into a unique point forecast. Options: mean, median, quantile, exponential

train(data, **kwargs)[source]

Method specific parameter fitting

Parameters
  • data – training time series data

  • kwargs – Method specific parameters

class pyFTS.models.ensemble.ensemble.SimpleEnsembleFTS(**kwargs)[source]

Bases: pyFTS.models.ensemble.ensemble.EnsembleFTS

An homogeneous FTS method ensemble with variations on partitionings and orders.

method

FTS method class that will be used on internal models

orders

Possible variations of order on internal models

partitioner_method

UoD partitioner class that will be used on internal methods

partitions

Possible variations of number of partitions on internal models

train(data, **kwargs)[source]

Method specific parameter fitting

Parameters
  • data – training time series data

  • kwargs – Method specific parameters

pyFTS.models.ensemble.ensemble.sampler(data, quantiles, bounds=False)[source]

pyFTS.models.ensemble.multiseasonal module

Silva, P. C. L et al. Probabilistic Forecasting with Seasonal Ensemble Fuzzy Time-Series XIII Brazilian Congress on Computational Intelligence, 2017. Rio de Janeiro, Brazil.

class pyFTS.models.ensemble.multiseasonal.SeasonalEnsembleFTS(name, **kwargs)[source]

Bases: pyFTS.models.ensemble.ensemble.EnsembleFTS

forecast_distribution(data, **kwargs)[source]

Probabilistic forecast one step ahead

Parameters
  • data – time series data with the minimal length equal to the max_lag of the model

  • kwargs – model specific parameters

Returns

a list with probabilistic.ProbabilityDistribution objects representing the forecasted Probability Distributions

train(data, **kwargs)[source]

Method specific parameter fitting

Parameters
  • data – training time series data

  • kwargs – Method specific parameters

update_uod(data)[source]
pyFTS.models.ensemble.multiseasonal.train_individual_model(partitioner, train_data, indexer)[source]

Module contents

Meta FTS that aggregates other FTS methods