Hi: I'm not at all familiar with the r-inla package so I can't help you
there. But any arima model can
be re-cast into its state space equivalent form. ( I think Jeremy Penzer
wrote a paper for showing how this is done in general ) So, one way would
be to convert the arima (1,0,1) model for example to it's state space
equivalent and then use say the DLM package like you mentioned. ( there are
others also like sspir etc but I'm not familar with them ) .
The DLM package is essentially the full R implementation of the theory in
West and Harrison" Bayesian Forcasting and Dynamic Models" text. So,
if
you're not familiar with that book, I would read it or atleast part of it
beeore getting involved with the DLM package. It's quite a nice package but
I think it's better if you read West and Harrison first. There may even be
a function in the DLM package for converting from arima to state space. I'm
not 100% sure about that but there probably is.
I hope that helps a little.
On Thu, Feb 14, 2013 at 4:27 PM, Juan Manuel Becerra
<jmbluengo@gmail.com>wrote:
> I'm searching a method to estimate the hyper-parameters in arima
models.
> I'm reading about r-inla package, but in the examples section only talk
> about the AR part of the arima, but i need help about the MA part too.
>
> I'm beginner in Bayesian methods, I'm reading the documentation
about dlm
> package and kalman filters, but the computacional cost of inla i think is
> better, but only AR models are supported for the moment. What's
packages
> can help me for that?.
>
> Someone can help me about it?. I need ligth for this question. If it's
> possible with a arima(1,0,1) example, w
>
> Thanks All
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@r-project.org mailing list
> 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]]