[R] plotting predicted values for time series.

Erin Hodgess erinm.hodgess at gmail.com
Mon Jun 7 17:09:00 CEST 2010

Dear R People:

I found the following code on the Internet:

> x <- co2
> m <- HoltWinters(x, alpha=.1, beta=0, gamma=0)
> p <- predict(m, n.ahead=240, prediction.interval=T)
> plot(m, predicted.values=p)

and it works fine.

My question:  is there an equivalent for ARIMA objects, please?  When
I tried, I got the following:
> x.arima <- arima(x,order=c(1,0,0))
> p1 <- predict(x.arima,n.ahead=6,prediction.interval=T)
> plot(x.arima,predicted.values=p1)
Error in xy.coords(x, y, xlabel, ylabel, log) :
  'x' is a list, but does not have components 'x' and 'y'
> plot(x,predicted.values=p1)
Warning messages:
1: In plot.window(xlim, ylim, log, ...) :
  "predicted.values" is not a graphical parameter
2: In title(main = main, xlab = xlab, ylab = ylab, ...) :
  "predicted.values" is not a graphical parameter
3: In axis(1, ...) : "predicted.values" is not a graphical parameter
4: In axis(2, ...) : "predicted.values" is not a graphical parameter
5: In box(...) : "predicted.values" is not a graphical parameter

Thanks in advance for any help!


Erin Hodgess
Associate Professor
Department of Computer and Mathematical Sciences
University of Houston - Downtown
mailto: erinm.hodgess at gmail.com

More information about the R-help mailing list