How to transform a discrete state space to a continuous state space Markov chain?

Discussion in 'Scientific Statistics Math' started by Kevin, Dec 17, 2003.

  1. Kevin

    Kevin Guest

    hi, everyone

    In one word, what we need is to transform a discrete state space MC to a
    continuous state space MC. Why? Because when the state space grows,
    the transition matrix will become terribly large. So it is very hard to
    obtain
    the invariant distribution Pi. The transition probability is just a
    multinormial
    distribution. I am just thinking to transform it to a continuous state
    space
    so that to get a continuous function of Pi. We already prove that the
    discrete
    state space MC has an invariant distribution. Does it also have a Pi
    function
    for the continous state space counterpart? Or under what conditions can
    we find this Pi function?

    Thanks for your help,

    Fan
     
    Kevin, Dec 17, 2003
    #1
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.