sarima.for: ARIMA Forecasting

In astsa: Applied Statistical Time Series Analysis

ARIMA Forecasting

Description

ARIMA forecasting.

Usage

sarima.for(xdata, n.ahead, p, d, q, P=0, D=0, Q=0, S=-1, tol = sqrt(.Machine$double.eps),
         no.constant = FALSE, plot = TRUE, plot.all = FALSE,  ylab = NULL, xreg = NULL, 
         newxreg = NULL, fixed = NULL, pcol = 2, pch = 1, ...)

Arguments

xdata	univariate time series
n.ahead	forecast horizon (number of periods)
p	AR order
d	difference order
q	MA order
P	SAR order; use only for seasonal models
D	seasonal difference; use only for seasonal models
Q	SMA order; use only for seasonal models
S	seasonal period; use only for seasonal models
tol	controls the relative tolerance (reltol) used to assess convergence. The default is sqrt(.Machine$double.eps), the R default.
no.constant	controls whether or not a constant is included in the model. If no.constant=TRUE, no constant is included in the model. See sarima for more details.
plot	if TRUE (default) the data (or some of it) and the forecasts and bounds are plotted
plot.all	if TRUE, all the data are plotted in the graphic; otherwise, only the last 100 observations are plotted in the graphic.
ylab	if NULL (default), the y-axis label is the name of the series.
xreg	Optionally, a vector or matrix of external regressors, which must have the same number of rows as the series. If this is used, newxreg MUST be specified.
newxreg	New values of xreg to be used for prediction. Must have at least n.ahead rows.
fixed	optional numeric vector of the same length as the total number of parameters. If supplied, only parameters corresponding to NA entries will be estimated.
pcol	color of the predictions in the graphic.
pch	plot character for the graphic. If plot.all=TRUE, this only applies to the forecasts.
...	additional graphical arguments

Details

For example, sarima.for(x, 5, 1,0,1) or sarima.for(x, 5, p=1, q=1) will forecast five time points ahead for an ARMA(1,1) fit to x. The output prints the forecasts and the standard errors of the forecasts, and supplies a graphic of the forecast with +/- 1 and 2 prediction error bounds.

Value

pred	the forecasts
se	the prediction (standard) errors

Missing Data

Yes it's ok if input as NA and the observations are vector or ts objects (meaning equally spaced). In this case, the graphic includes a line with points. Otherwise, lone observations would not be visible.

Note

If plot.all=TRUE, the data are displayed as a line only unless there are missing observations; see the Missing Data section. Points (and more) can be added to the graphic as long as the device stays open. For example:

sarima.for(gtemp_land, 10, d=1, q=1, plot.all=TRUE, pch=19)
points(gtemp_land, pch=20, col=4)
abline(v=2024, col=6, lty=5)
text(2000, 2.2, "it's getting hot in here", font=2, col=6, srt=45)

References

You can find demonstrations of astsa capabilities at FUN WITH ASTSA.

The most recent version of the package can be found at https://github.com/nickpoison/astsa/.

In addition, the News and ChangeLog files are at https://github.com/nickpoison/astsa/blob/master/NEWS.md.

The webpages for the texts and some help on using R for time series analysis can be found at https://nickpoison.github.io/.

See Also

sarima

Examples

sarima.for(gtemp_both, n.ahead=6, d=1, q=1, col=6, pcol=4, gg=TRUE)

# with regressors 
nummy   = length(soi)
n.ahead = 24 
nureg   = time(soi)[nummy] + seq(1,n.ahead)/12
sarima.for(soi, n.ahead, 2,0,0, 2,0,0,12, xreg=time(soi), newxreg=nureg) 

# missing data
sarima.for(ar1miss, n.ahead=5, p=1, pch=19)

astsa documentation built on Aug. 21, 2025, 5:47 p.m.