[R] Markov chain simulation

[Berend Hasselman]

On 28-01-2014, at 19:23, Kaptue Tchuente, Armel <armel.kaptue at sdstate.edu> wrote:

> Hi Kehl,
> 
> Thank you for your reply.
> 
> Indeed, there are many ways to simulate a markov chain sequence.
> 
> For instance, if I assume that n=20, rnd_occ1<-c(0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1) and rnd_occ2<-c(0, 0, 0, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0) are two markov chain sequences having the transition matrix TransitionMatrix<- matrix(c(0.7, 0.3, 0.4, 0.6),byrow=TRUE, nrow=2)
> For the first sequence the state "1" has 9 occurences while for the second sequence, the state "1" has 6 occurences.
> What I'm looking for it is an algorithm or a library to simulate efficiently such a markov chain sequence with for instance 12 ooccurences of the state ‘1'
> 

I used package sos to search for Markov  (using library(sos); findFn(“Markov”).
Amongst others this what I found which seem to be appropriate:

markovchain
MCMCpack
DTMCPack

and a lot more.
BTW: it’s no use asking me more about Markov. I am ignorant in that subject.

Good luck,

Berend