[R] FW: Multivariate ARIMA

[Thomas Lofaro]

Hi Paul,

Thanks so much for reaching out.  

I read through the link but I am not having any luck.  It seems the first part of the link refers to a single variable ARIMA model with no exogenous variable.  Then the link starts discussing forecasting.

Anyway, I have two columns of data in my matrix (each row represents a new month).  I want column A to predict Column B, where I can adjust the lag to be say 1,2, or 6 months.

Is there any example R code that accomplishes this that you know of?  Apologies for my lack of proficiency in the program.


-----Original Message-----
From: Paul Gilbert [mailto:pgilbert902 at gmail.com] 
Sent: Monday, February 15, 2016 6:27 PM
To: Thomas Lofaro
Cc: r-help at r-project.org
Subject: Re: [R] FW: Multivariate ARIMA

See also package dse.  There are examples in the guide: 
http://cran.at.r-project.org/web/packages/dse/vignettes/Guide.pdf

Paul

On 02/14/2016 06:00 AM, r-help-request at r-project.org wrote:
> Date: Fri, 12 Feb 2016 18:12:37 +0000 From: Thomas 
> Lofaro<Thomas.Lofaro at SurveySampling.com> To:"r-help at R-project.org"
> <r-help at R-project.org> Subject: [R] FW: Multivariate ARIMA 
> Question.... Message-ID:
> <BN4PR07MB22103CC8EB3FF13CCDF00D4D93A90 at BN4PR07MB2210.namprd07.prod.ou
> tlook.com>
>
>  Content-Type: text/plain; charset="UTF-8"
>
> Hi, sorry, I will try to make this short. Essentially, what I am 
> trying to do is run a multivariate ARIMA model predicting one variable 
> with another.  However, the only R packages I have found to 
> accommodate such an analysis are ARIMAX and VAR.  Unfortunately, there 
> don?t seem to be any good tutorials or demonstrations of how to 
> actually perform this analysis in practice, on my two columns of data.  
> I was wondering if anyone knew of a resource that might help (or one 
> that provided an example of r code to accomplish this).
> Thanks sooo much!