Skip to content
Back to formatted view

Raw Message

Message-ID: <CAMi=pg7hYX7KM_1wXEpOYE4phuzmnCGuEn4Fs5cj1R+hkG2pLQ@mail.gmail.com>
Date: 2013-08-29T01:42:29Z
From: Whit Armstrong
Subject: rcppbugs and markov switching model
In-Reply-To: <CAF6gDEWnFdJ337zc8=7A0XRrXuTYCxscvqSNbJ=rQdZM8feb4A@mail.gmail.com>

Have you thought about trying viterbi first?

There is an example here:
http://www.r-bloggers.com/regime-detection/

If you don't want to use that you can try the example here on about
page page 29:
http://www.rinfinance.com/agenda/2012/talk/WhitArmstrong.pdf

If you decide you want to use mcmc, I think the way to go is to write
the model in c++ and call it with Rcpp.

You should override the random walk classes for bernoulli and use
gibbs within metropolis.

There is an example of that here (for a linear model):
https://github.com/armstrtw/CppBugs/tree/master/test/r.inline.example

If you need some help then feel free to contact me off list.

-Whit


On Wed, Aug 28, 2013 at 9:10 PM, pu chen <chenpuias at gmail.com> wrote:
> Anyone know how to use rcppbugs to estimate markov switching model? Thanks.
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-SIG-Finance at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only. If you want to post, subscribe first.
> -- Also note that this is not the r-help list where general R questions should go.