mchain {HiddenMarkov}R Documentation

Markov Chain Object

Description

Creates a Markov chain object with class "mchain". It does not simulate data.

Usage

mchain(x, Pi, delta, nonstat = TRUE)

Arguments

x

is a vector of length nn containing the observed process, else it is specified as NULL. This is used when there are no data and a process is to be simulated.

Pi

is the m×mm \times m transition probability matrix of the Markov chain.

delta

is the marginal probability distribution of the mm state Markov chain at the first time point.

nonstat

is logical, TRUE if the homogeneous Markov chain is assumed to be non-stationary, default. See “Details” below.

Value

A list object with class "mchain", containing the above arguments as named components.

Examples

Pi <- matrix(c(0.8, 0.2,
               0.3, 0.7),
             byrow=TRUE, nrow=2)

#    Create a Markov chain object with no data (NULL)
x <- mchain(NULL, Pi, c(0,1))

#    Simulate some data
x <- simulate(x, nsim=2000)

#   estimate transition probabilities
estPi <- table(x$mc[-length(x$mc)], x$mc[-1])
rowtotal <- estPi %*% matrix(1, nrow=nrow(Pi), ncol=1)
estPi <- diag(as.vector(1/rowtotal)) %*% estPi
print(estPi)

[Package HiddenMarkov version 1.8-13 Index]