forecastHybrid5.0.19 package

Convenient Functions for Ensemble Time Series Forecasts

accuracy.cvts

Accuracy measures for cross-validated time series

accuracy.hybridModel

Accuracy measures for hybridModel objects

checkCVArguments

Validate that CV window parameters are valid

checkModelArgs

Helper function to test all the model arguments (e.g. a.args, e.args, ...

checkParallelArguments

Helper function to check the that the parallel arguments are valid

cvts

Cross validation for time series

extractForecasts

Extract cross validated rolling forecasts

fitted.hybridModel

Extract Model Fitted Values

forecast.hybridModel

Hybrid forecast

forecast.thetam

Forecast using a Theta model

getModel

Return a forecast model function for a given model character

getModelName

Translate character to model name

hybridModel

Hybrid time series modeling

is.hybridModel

Test if the object is a hybridModel object

plot.hybridModel

Plot a hybridModel object

plot.thetam

Plot components from Theta model

plotFitted

Plot the fitted values of a hybridModel object

plotModelObjects

Plot the component models of a hybridModel object

prepareTimeseries

Helper function to validate and clean the input time series

print.hybridModel

Print information about the hybridModel object

removeModels

Helper function to remove models that require more data

residuals.hybridModel

Extract Model Residuals

summary.hybridModel

Print a summary of the hybridModel object

thetam

Theta method 'model'

thiefModel

Forecast ensemble using THieF

tsCombine

Combine multiple sequential time series

tsPartition

Generate training and test indices for time series cross validation

tsSubsetWithIndices

Subset time series with provided indices

unwrapParallelModels

Helper function used to unpack the fitted model objects from a list

Convenient functions for ensemble forecasts in R combining approaches from the 'forecast' package. Forecasts generated from auto.arima(), ets(), thetaf(), nnetar(), stlm(), tbats(), and snaive() can be combined with equal weights, weights based on in-sample errors (introduced by Bates & Granger (1969) <doi:10.1057/jors.1969.103>), or cross-validated weights. Cross validation for time series data with user-supplied models and forecasting functions is also supported to evaluate model accuracy.

  • Maintainer: David Shaub
  • License: GPL-3
  • Last published: 2020-08-28