Package: seer 1.1.8

seer: Feature-Based Forecast Model Selection

A novel meta-learning framework for forecast model selection using time series features. Many applications require a large number of time series to be forecast. Providing better forecasts for these time series is important in decision and policy making. We propose a classification framework which selects forecast models based on features calculated from the time series. We call this framework FFORMS (Feature-based FORecast Model Selection). FFORMS builds a mapping that relates the features of time series to the best forecast model using a random forest. 'seer' package is the implementation of the FFORMS algorithm. For more details see our paper at <https://www.monash.edu/business/econometrics-and-business-statistics/research/publications/ebs/wp06-2018.pdf>.

Authors:Thiyanga Talagala [aut, cre], Rob J Hyndman [ths, aut], George Athanasopoulos [ths, aut]

seer_1.1.8.tar.gz
seer_1.1.8.zip(r-4.5)seer_1.1.8.zip(r-4.4)seer_1.1.8.zip(r-4.3)
seer_1.1.8.tgz(r-4.4-any)seer_1.1.8.tgz(r-4.3-any)
seer_1.1.8.tar.gz(r-4.5-noble)seer_1.1.8.tar.gz(r-4.4-noble)
seer_1.1.8.tgz(r-4.4-emscripten)seer_1.1.8.tgz(r-4.3-emscripten)
seer.pdf |seer.html
seer/json (API)

# Install 'seer' in R:
install.packages('seer', repos = c('https://thiyangt.r-universe.dev', 'https://cloud.r-project.org'))

Peer review:

Bug tracker:https://github.com/thiyangt/seer/issues

On CRAN:

5.32 score 80 stars 52 scripts 191 downloads 1 mentions 37 exports 61 dependencies

Last updated 2 years agofrom:abd4c2aa17. Checks:OK: 7. Indexed: yes.

TargetResultDate
Doc / VignettesOKOct 31 2024
R-4.5-winOKOct 31 2024
R-4.5-linuxOKOct 31 2024
R-4.4-winOKOct 31 2024
R-4.4-macOKOct 31 2024
R-4.3-winOKOct 31 2024
R-4.3-macOKOct 31 2024

Exports:accuracy_arimaaccuracy_etsaccuracy_mstlaccuracy_nnaccuracy_rwaccuracy_rwdaccuracy_snaiveaccuracy_stlaraccuracy_tbatsaccuracy_thetaaccuracy_wnacf_seasonalDiffacf5build_rfcal_featurescal_m4measurescal_MASEcal_medianscaledcal_sMAPEcal_WAclassify_labelsclasslabelcombination_forecast_insideconvert_mstse_acf1fcast_accuracyfforms_combinationforecastfforms_ensembleholtWinter_parametersprepare_trainingsetrf_forecastsim_arimabasedsim_etsbasedsim_mstlbasedsplit_namesstlarunitroot

Dependencies:clicodetoolscolorspacecurldigestdplyrfansifarverforecastforecThetafracdifffurrrfuturegenericsggplot2globalsgluegtableisobandjsonlitelabelinglatticelifecyclelistenvlmtestmagrittrMASSMatrixmgcvmunsellnlmennetparallellypillarpkgconfigpurrrquadprogquantmodR6randomForestRColorBrewerRcppRcppArmadilloRcppRollrlangscalesstringistringrtibbletidyselecttimeDatetseriestsfeaturesTTRurcautf8vctrsviridisLitewithrxtszoo

Readme and manuals

Help Manual

Help pageTopics
Calculate accuracy measue based on ARIMA modelsaccuracy_arima
Forecast-accuracy calculationaccuracy_ets
Calculate accuracy based on MSTLaccuracy_mstl
Calculate accuracy measure calculated based on neural network forecastsaccuracy_nn
Calculate accuracy measure based on random walk modelsaccuracy_rw
Calculate accuracy measure based on random walk with driftaccuracy_rwd
Calculate accuracy measure based on snaive methodaccuracy_snaive
Calculate accuracy measure based on STL-AR methodaccuracy_stlar
Calculate accuracy measure based on TBATSaccuracy_tbats
Calculate accuracy measure based on Theta methodaccuracy_theta
Calculate accuracy measure based on white noise processaccuracy_wn
Autocorrelation coefficients based on seasonally differenced seriesacf_seasonalDiff
Autocorrelation-based featuresacf5
build random forest classifierbuild_rf
Calculate features for new time series instancescal_features
Mean of MASE and sMAPEcal_m4measures
Mean Absolute Scaled Error(MASE)cal_MASE
scale MASE and sMAPE by mediancal_medianscaled
symmetric Mean Absolute Pecentage Error(sMAPE)cal_sMAPE
Weighted Averagecal_WA
Classify labels according to the FFORMS fameworkclassify_labels
identify the best forecasting methodclasslabel
This function is call to be inside fforms_combinationcombination_forecast_inside
Convert multiple frequency time series into msts objectconvert_msts
Autocorrelation coefficient at lag 1 of the residualse_acf1
calculate forecast accuracy from different forecasting methodsfcast_accuracy
Combination forecast based on fformsfforms_combinationforecast
Function to identify models to compute combination forecast using FFORMS algorithmfforms_ensemble
Parameter estimates of Holt-Winters seasonal methodholtWinter_parameters
preparation of training setprepare_trainingset
function to calculate point forecast, 95% confidence intervals, forecast-accuracy for new seriesrf_forecast
Simulate time series based on ARIMA modelssim_arimabased
Simulate time series based on ETS modelssim_etsbased
Simulate time series based on multiple seasonal decompositionsim_mstlbased
split the names of ARIMA and ETS modelssplit_names
STL-AR methodstlar
Unit root test statisticsunitroot