getStationaryDistribution: getStationaryDistribution

In qiwei-li/infiniteMC: Finite and Infinite, Discrete and Continuous markov chain toolkit

Description

Calculate the stationary distribution of a markov chain if it exists

Usage

getStationaryDistribution(mc.obj, epsilon=0.01, iteration=30, totalProb=0.9)

Arguments

mc.obj	Markov chain object created by the 'mc.create' function
epsilon	The threshold for the converging (in cosine). Needed in infinite markov chain. Default to 0.01.
iteration	The max iteration for the approximation algorithm. Needed in infinite markov chain. Default to 30
totalProb	The proportion of information that the approximate distribution contains. Default to 0.9

Value

In the case of finite, returns the stationary distribution. In the case of infinite, returns a list containing an success indicator, converge state count, approximated stationary distribution, and converge transition matrix.

Examples

singleServer = function(i,j){
  if(i==1 && j==2)
    return(1)
  p = 0.3
  q = 0.7
  r = 0
  if(j == i+1)
    return(p)
  if(j == i-1)
    return(q)
  if(j==i)
    return(r)
  return(0)
}
ex = mc.create(pijdef=singleServer, discrete=TRUE, infinite=TRUE)
ans = getStationaryDistribution(ex)

qiwei-li/infiniteMC documentation built on May 26, 2019, 11:36 a.m.