Autoplot time series models (like AR, ARIMA)

Usage

# S3 method for class 'tsmodel'
autoplot(
  object,
  data = NULL,
  predict = NULL,
  is.date = NULL,
  ts.connect = TRUE,
  fitted.geom = "line",
  fitted.colour = "#FF0000",
  fitted.size = NULL,
  fitted.linetype = NULL,
  fitted.alpha = NULL,
  fitted.fill = NULL,
  fitted.shape = NULL,
  predict.geom = "line",
  predict.colour = "#0000FF",
  predict.size = NULL,
  predict.linetype = NULL,
  predict.alpha = NULL,
  predict.fill = NULL,
  predict.shape = NULL,
  conf.int = TRUE,
  conf.int.colour = "#0000FF",
  conf.int.linetype = "none",
  conf.int.fill = "#000000",
  conf.int.alpha = 0.3,
  ...
)

Arguments

object: Time series model instance
data: original dataset, needed for stats::ar, stats::Arima
predict: Predicted stats::ts If not provided, try to retrieve from current environment using variable name.
is.date: Logical frag indicates whether the stats::ts is date or not. If not provided, regard the input as date when the frequency is 4 or 12
ts.connect: Logical frag indicates whether connects original time-series and predicted values
fitted.geom: geometric string for fitted time-series
fitted.colour: line colour for fitted time-series
fitted.size: point size for fitted time-series
fitted.linetype: line type for fitted time-series
fitted.alpha: alpha for fitted time-series
fitted.fill: fill colour for fitted time-series
fitted.shape: point shape for fitted time-series
predict.geom: geometric string for predicted time-series
predict.colour: line colour for predicted time-series
predict.size: point size for predicted time-series
predict.linetype: line type for predicted time-series
predict.alpha: alpha for predicted time-series
predict.fill: fill colour for predicted time-series
predict.shape: point shape for predicted time-series
conf.int: Logical flag indicating whether to plot confidence intervals
conf.int.colour: line colour for confidence intervals
conf.int.linetype: line type for confidence intervals
conf.int.fill: fill colour for confidence intervals
conf.int.alpha: alpha for confidence intervals
...: Keywords passed to autoplot.ts

Value

ggplot

Examples

if (FALSE) { # \dontrun{
d.ar <- stats::ar(AirPassengers)
autoplot(d.ar)
autoplot(d.ar, predict = predict(d.ar, n.ahead = 5))
autoplot(stats::arima(UKgas), data = UKgas)
autoplot(forecast::arfima(AirPassengers))
autoplot(forecast::nnetar(UKgas), is.date = FALSE)

d.holt <- stats::HoltWinters(USAccDeaths)
autoplot(d.holt)
autoplot(d.holt, predict = predict(d.holt, n.ahead = 5))
autoplot(d.holt, predict = predict(d.holt, n.ahead = 5, prediction.interval = TRUE))
} # }