predict.HoltWinters {ts}R Documentation

prediction function for fitted Holt-Winters models

Description

Computes predictions and prediction intervals for models fitted by the Holt-Winters method.

Usage

predict(object, n.ahead=1, prediction.interval = FALSE,
                    quantile = qnorm(0.975), ...)

Arguments

object An object of class HoltWinters
n.ahead Number of future periods to predict
prediction.interval logical. If TRUE, the lower and upper bounds of the corresponding prediction intervals are computed
quantile Quantile used for each one-sided prediction interval (default: the 0.975 quantile to give a two-sided 0.95 interval)
... arguments passed to or from other methods.

Value

A time series of the predicted values. If prediction intervals are requested, a multiple time series is returned with columns fit, lwr and upr for the predicted values and the lower and upper bounds respectively.

Author(s)

David Meyer david.meyer@ci.tuwien.ac.at

References

C.C Holt (1957) Forecasting seasonals and trends by exponentially weighted moving averages, ONR Research Memorandum, Carnigie Institute 52.

P.R Winters (1960) Forecasting sales by exponentially weighted moving averages, Management Science 6, 324–342.

See Also

HoltWinters

Examples

data(co2)
m <- HoltWinters(co2)
p <- predict(m, 50, prediction.interval = TRUE)
plot(m, p)

[Package Contents]