[R] Arma - estimate of variance of white noise variables

[Rui Barradas]

Hello,

Em 29-04-2013 13:49, Preetam Pal escreveu:
> Hi all,
>
> Suppose I am fitting an arma(p,q) model to a time series y_t.
> So, my model should contain (q+1) white noise variables.

Why? How on hearth can you say this?

> As far as I know, each of them should have the same variance.
> How do I get the estimate of this variance by running the arma(y) function
> (or is there any other way)?

I'm not certain that the following is what you're looking for.

library(tseries)
fit <- arma(y, ...etc...)
var(resid(fit))


Hope this helps,

Rui Barradas
>
> Appreciate your help.
>
> Thanks,
> Preetam
>