# forecast v7.3

0

Monthly downloads

by Rob Hyndman

## Forecasting Functions for Time Series and Linear Models

Methods and tools for displaying and analysing
univariate time series forecasts including exponential smoothing
via state space models and automatic ARIMA modelling.

## Readme

The R package *forecast* provides methods and tools for displaying and analysing univariate time series forecasts including exponential smoothing via state space models and automatic ARIMA modelling.

## Installation

You can install the **stable** version on
R CRAN.

```
install.packages('forecast', dependencies = TRUE)
```

You can install the **development** version from
Github

```
# install.packages("devtools")
devtools::install_github("robjhyndman/forecast")
```

## Usage

```
library(forecast)
# ETS forecasts
fit <- ets(USAccDeaths)
plot(forecast(fit))
# Automatic ARIMA forecasts
fit <- auto.arima(WWWusage)
plot(forecast(fit, h=20))
# ARFIMA forecasts
library(fracdiff)
x <- fracdiff.sim( 100, ma=-.4, d=.3)$series
fit <- arfima(x)
plot(forecast(fit, h=30))
# Forecasting with STL
tsmod <- stlm(USAccDeaths, modelfunction=ar)
plot(forecast(tsmod, h=36))
plot(stlf(AirPassengers, lambda=0))
decomp <- stl(USAccDeaths,s.window="periodic")
plot(forecast(decomp))
# TBATS forecasts
fit <- tbats(USAccDeaths)
plot(forecast(fit))
taylor.fit <- tbats(taylor)
plot(forecast(taylor.fit))
```

## License

This package is free and open source software, licensed under GPL (>= 2).

## Functions in forecast

Name | Description | |

autoplot.decomposed.ts | ggplot of a decomposed time series object | |

arima.errors | ARIMA errors | |

autoplot.acf | ggplot (Partial) Autocorrelation and Cross-Correlation Function Estimation | |

auto.arima | Fit best ARIMA model to univariate time series | |

accuracy | Accuracy measures for forecast model | |

arimaorder | Return the order of an ARIMA or ARFIMA model | |

arfima | Fit a fractionally differenced ARFIMA model | |

Acf | (Partial) Autocorrelation and Cross-Correlation Function Estimation | |

Arima | Fit ARIMA model to univariate time series | |

autoplot.stl | ggplot STL object | |

CV | Cross-validation statistic | |

dm.test | Diebold-Mariano test for predictive accuracy | |

bizdays | Number of trading days in each season | |

dshw | Double-Seasonal Holt-Winters Forecasting | |

BoxCox | Box Cox Transformation | |

autoplot.ts | Automatically create a ggplot for time series objects | |

easter | Easter holidays in each season | |

BoxCox.lambda | Automatic selection of Box Cox transformation parameter | |

bats | BATS model (Exponential smoothing state space model with Box-Cox transformation, ARMA errors, Trend and Seasonal components) | |

croston | Forecasts for intermittent demand using Croston's method | |

fitted.tbats | h-step in-sample forecasts using tbats models | |

fitted.Arima | h-step in-sample forecasts using ARIMA models | |

ets | Exponential smoothing state space model | |

forecast.ets | Forecasting using ETS models | |

fitted.bats | h-step in-sample forecasts using bats models | |

findfrequency | Find dominant frequency of a time series | |

forecast.bats | Forecasting using BATS and TBATS models | |

fitted.ets | h-step in-sample forecasts using ets models | |

forecast.Arima | Forecasting using ARIMA or ARFIMA models | |

fitted.nnetar | h-step in-sample forecasts using nnetar models | |

forecast.StructTS | Forecasting using Structural Time Series models | |

forecast.mlm | Forecast a multiple linear model with possible time series components | |

forecast.HoltWinters | Forecasting using Holt-Winters objects | |

forecast.lm | Forecast a linear model with possible time series components | |

gas | Australian monthly gas production | |

forecast.stl | Forecasting using stl objects | |

forecast.nnetar | Forecasting using neural network models | |

fortify.forecast | Fortify a forecast object to data.frame for ggplot | |

plot.forecast | Forecast plot | |

forecast | Forecasting time series | |

ggmonthplot | Create a seasonal subseries ggplot | |

gglagplot | Time series lag ggplots | |

geom_forecast | Forecast plot | |

is.constant | Is an object constant? | |

gold | Daily morning gold prices | |

is.ets | Is an object a particular model type? | |

na.interp | Interpolate missing values in a time series | |

naive | Naive and Random Walk Forecasts | |

is.forecast | Is an object a particular forecast type? | |

getResponse | Get response variable from time series model. | |

ma | Moving-average smoothing | |

logLik.ets | Log-Likelihood of an ets object | |

plot.Arima | Plot characteristic roots from ARIMA model | |

wineind | Australian total wine sales | |

seasonaldummy | Seasonal dummy variables | |

plot.bats | Plot components from BATS model | |

seasadj | Seasonal adjustment | |

woolyrnq | Quarterly production of woollen yarn in Australia | |

sindexf | Forecast seasonal index | |

simulate.ets | Simulation from a time series model | |

taylor | Half-hourly electricity demand | |

tbats.components | Extract components of a TBATS model | |

plot.ets | Plot components from ETS model | |

plot.mforecast | Multivariate forecast plot | |

tsoutliers | Identify and replace outliers in a time series | |

monthdays | Number of days in each season | |

tslm | Fit a linear model with time series components | |

msts | Multi-Seasonal Time Series | |

subset.ts | Subsetting a time series | |

splinef | Cubic Spline Forecast | |

tbats | TBATS model (Exponential smoothing state space model with Box-Cox transformation, ARMA errors, Trend and Seasonal components) | |

thetaf | Theta method forecast | |

tsclean | Identify and replace outliers and missing values in a time series | |

ndiffs | Number of differences required for a stationary series | |

tsdisplay | Time series display | |

nnetar | Neural Network Time Series Forecasts | |

mforecast | Forecasting time series | |

seasonplot | Seasonal plot | |

meanf | Mean Forecast | |

ses | Exponential smoothing forecasts | |

No Results! |

## Last month downloads

## Details

LinkingTo | Rcpp (>= 0.11.0), RcppArmadillo (>= 0.2.35) |

LazyData | yes |

ByteCompile | TRUE |

BugReports | https://github.com/robjhyndman/forecast/issues |

License | GPL (>= 2) |

URL | http://github.com/robjhyndman/forecast |

NeedsCompilation | yes |

Packaged | 2016-10-12 10:12:00 UTC; robjhyndman |

Repository | CRAN |

Date/Publication | 2016-10-13 00:38:06 |

imports | colorspace , fracdiff , ggplot2 (>= 2.0.0) , nnet , parallel , Rcpp (>= 0.11.0) , tseries |

depends | graphics , R (>= 3.0.2) , stats , timeDate , zoo |

linkingto | RcppArmadillo (>= 0.2.35) |

suggests | testthat |

Contributors | Rob Hyndman |

#### Include our badge in your README

```
[![Rdoc](http://www.rdocumentation.org/badges/version/forecast)](http://www.rdocumentation.org/packages/forecast)
```