[R] R Arimax Function

Paul Bernal paulbernal07 at gmail.com
Thu Dec 4 06:12:43 CET 2014

Hello Mark,

Thank you for your timely reply. All I want to know is if the xreg argument
must contain the same amount of historical observations as the dependent
variable and if newxreg  has to have a forecast of the exogenous variable
equal to the number of periods to be forecasted.

Best regards and thanks again,


2014-12-03 22:30 GMT-06:00 Mark Leeds <markleeds2 at gmail.com>:

> hi paul: I don't have time to answer all of your questions but below
> should help.
> there'sa  better one that actually how arima is better for forecasting
> than gls
> but I can't find it right now. if I find it, I'll send it.
> http://stats.stackexchange.com/questions/6469/simple-linear-model-with-autocorrelated-errors-in-r
> On Wed, Dec 3, 2014 at 11:12 PM, Paul Bernal <paulbernal07 at gmail.com>
> wrote:
>> Hello everyone,
>> I am just trying to understand how the Arimax function works, so my
>> questions are:
>> 1. If I have a univariate time series of sales and sales are dependent
>> upon, say inflation rates, then my xreg would be inflation rates right?
>> 2. Now is I have historial data on sales (from january 2000 to december
>> 2010) then my xreg argument would be a historical time series of inflation
>> rates on the same time frame? (from january 2000 to december 2010?)
>> 3. If I want to predit, say 5 years of monthly data (60 periods) using the
>> exogenous variable (in this particular case inflation rates) then that
>> means I would have to have a a 60 period forecast of inflation rates and
>> that would be my newxreg argument?
>> Any guidance will be greatly appreciated,
>> Best regards,
>> Paul
>>         [[alternative HTML version deleted]]
>> ______________________________________________
>> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
>> http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.

	[[alternative HTML version deleted]]

More information about the R-help mailing list