Movatterモバイル変換

fable

The R packagefable provides a collection of commonly usedunivariate and multivariate time series forecasting models includingexponential smoothing via state space models and automatic ARIMAmodelling. These models work within the fable framework, which providesthe tools to evaluate, visualise, and combine models in a workflowconsistent with the tidyverse.

Installation

You can install thestable version fromCRAN:

install.packages("fable")

You can install thedevelopment version fromGitHub

# install.packages("remotes")remotes::install_github("tidyverts/fable")

Installing this software requires a compiler

Example

library(fable)library(tsibble)library(tsibbledata)library(lubridate)library(dplyr)aus_retail%>%filter(    State%in%c("New South Wales","Victoria"),    Industry=="Department stores"  )%>%model(ets =ETS(box_cox(Turnover,0.3)),arima =ARIMA(log(Turnover)),snaive =SNAIVE(Turnover)  )%>%forecast(h ="2 years")%>%autoplot(filter(aus_retail,year(Month)>2010),level =NULL)

Learning to forecast withfable

• The pkgdown site describes all models provided by fable, and howthey are used:https://fable.tidyverts.org/
• The forecasting principles and practices online textbook provides anintroduction to time series forecasting using fable:https://otexts.com/fpp3/(WIP)

Getting help

• Questions about forecasting can be asked onCrossValidated.

• Common questions about the fable package are often found onStack Overflow. Youcan use this to ask for help if the question isn’t already answered. Aminimally reproducibleexample that describes your issue is the best way to ask forhelp!