[R-sig-Geo] categorical AR or ARMA time series processes

[Freddy López]

Hello Emanuele,

Perhaps GARMA models can be useful. A reference:

http://pubs.amstat.org/doi/abs/10.1198/016214503388619238

This is implemented (partially, I think) in VGAM library. See garma()
function. I could model some years ago count data using an
autoregressive framework with this package.

I hope this help.

Cheers.

On Fri, Jun 24, 2011 at 05:27, Emanuele Cordano
<emanuele.cordano at gmail.com> wrote:
>
> Dear all,
>
> I'm working on daily precipitation generation time series, in particular I
> want to model a binary time serries (wet/dry day) for several years. I read
> about Discrete-Value or Catagorical-value AR or ARMA techniques. They refer
> to  Pegram’s [Pegram, G.G.S., 1980. An autoregressive model for multilag
> markov chains. J. Appl. Probab. 17, 350–362] mixing operator and subsequent
> more recent works  like
> http://www.sciencedirect.com/science/article/pii/S0167715209001977 or
> http://www.sciencedirect.com/science/article/pii/S0378375809000330.
> Are these methods and Pegram's operetor  already implemented in some CRAN
> package  or something similar? I'm looking if there is something already
> implemented on which I can work
>
> Thanks a lot in advance
>
> Regards
> Emanuele Cordano
>
>        [[alternative HTML version deleted]]
>
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>



--
«But Gwindor answered: 'The doom lies in yourself, not in your name.'»

JRR Tolkien