Re: How to interpret coefficients from ARMA in R (r-project.org)?

From: Marc Schwartz (MSchwartz_at_mn.rr.com)
Date: 06/13/04 

Date: Sun, 13 Jun 2004 15:37:52 GMT

On Sun, 13 Jun 2004 07:15:09 -0700, Bush will disarm all workers next
wrote:

> *** Startz <richardstartz1@comcast.net> wrote in message
> news:<m9qmc0tohr3u9or2t3b5h50g7ekfuhas6i@4ax.com>...
>
>> Out of curiosity, is the original data an asset price?
>
> The original data are peak traffic at a road intersection. In R-1.9.0,
> there is a predict function for ARIMA but not for ARMA. Assuming the
> model is stationary, how can one forecast/predict using the results of
> ARMA?
>
> Aso Tried this:
>> fit<-arima(inp, order=c(0,2,0))
> Error in "[<-"(`*tmp*`, 1, 1, value = 1/(1 - phi^2)) :
> nothing to replace with
>
> Thanks!
> Dakshin

I do not use ARMA/ARIMA, however a search of the r-help list archives and
the tseries package documentation seems to lead one to arima0, for which
there is a predict method. That is a zero, not the letter "O".

More information and examples are in ?arima0.

HTH,

Marc Schwartz