Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have a time series and fitted an AR(I)MA object (I really just want an ARMA process so the order is (p,0,q)) in R. Now I'd like to "continue" that process (I guess what I mean is perform simulations).

I found arima.sim() but I'm not sure how to get it to use my existing time series (that I'd like to "continue").

I hope my question makes sense!


share|improve this question
I think that the word you're looking for is forecasting. – johnny Dec 3 '12 at 8:14
fit <- Arima(x,order=c(p,0,q))
newx <- simulate(fit,nsim=100)
share|improve this answer
Thank you, that seems to be exactly what I was looking for! – Lucas Dec 3 '12 at 10:20
Is there a way to specify the distribution / standard deviation for the white noise process? – Lucas Dec 3 '12 at 10:31
The default is a normal distribution with sd equal to the sd of the residuals. You can change the latter by altering the value of fit$sigma2. It is also possible to use a bootstrap distribution of the residuals by setting bootstrap=TRUE. – Rob Hyndman Dec 3 '12 at 21:33
Thanks for the reply! How would I go about changing the distribution itself (say to a t distribution)? – Lucas Dec 4 '12 at 19:20
Write your own version of the simulate.Arima function. – Rob Hyndman Dec 4 '12 at 21:32

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.