[R] Revo R for Arima Implementation

Rolf Turner r.turner at auckland.ac.nz
Mon Oct 28 22:07:29 CET 2013

Your question still makes no sense at all to me.  You provide no example 
and no data.  You do not specify, in any comprehensible way, what model you
are trying to fit.

And as Jeff Newmiller said, if you are concerned with "Revo R" why are 
you posting
to "r-help"?

Please get your act together or stop cluttering up the r-help list.


     Rolf Turner

On 10/29/13 02:26, Anindita Chattopadhyay wrote:
> Hi Rolf,
> Thanks for the response. I have re-phrased the problem. Hope this will help.
> We're working on a project where our model tries to predict the value of bookings as a time series of past bookings as well as some external flags.
> The data was found to be stationary using the Dicky-Fuller test.
> We have only a non-seasonal component for the AR and MA terms (small p & q only).
> No differencing was done.
> Then, we used values of p = 1, 7, 8 and q = 1, 7.
> The main issue is that the p values are additive and q values are multiplicative.
> We need to understand how we can implement this in Revo R.
> Just to add, since we have external flags , we cannot make use of SARIMA function.
> Please let me know if any other questions.
> Thank you in advance!
> Regards,
> Anindita Chattopadhyay | +919886800606 | www.mu-sigma.com |
> -----Original Message-----
> From: Rolf Turner [mailto:r.turner at auckland.ac.nz]
> Sent: Saturday, October 26, 2013 1:56 AM
> To: Anindita Chattopadhyay
> Cc: r-help at r-project.org; Harish K
> Subject: Re: [R] Revo R for Arima Implementation
> Your question is pretty well totally opaque to me.  Describe the model you want to fit in mathematical terms, rather than referring to The Package That Must Not Be Named.  This is the ***R***  list.
> It is possible that you might want to make use of the "seasonal"
> argument to the
> arima() function.
>       cheers,
>       Rolf Turner
> On 10/26/13 01:19, Anindita Chattopadhyay wrote:
>> Hello There,
>> We have used ARIMA(multiplicative MA and additive AR) model in SAS to come to our results. Please let me know how could that be implemented in R.
>> Like,  In SAS we can pass the variable  as AR= (1,7) and MA=(1)(7) which is additive and multiplicative respectively.
>> In R we have the option as : order = c(0, 0, 0) which is (p,d,q).Is there any specific way/code such that we specify the additive  AR and multiplicative MA simultaneously?

More information about the R-help mailing list