1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Markov chain - stationary distribution - please help

  1. Apr 22, 2008 #1
    1. The problem statement, all variables and given/known data

    -A taxi is located either at the airport or in the city. From the city, the next trip is to the airport with 1/4 probability, or to somewhere else in the city with 3/4. From the airport, the next trip is always to the city.

    (a) find the stationary distribution
    (b) starting from the airport, what is the expected number of trips until its next visit to the airport?

    2. Relevant equations



    3. The attempt at a solution

    The matrix I got is

    0 1
    1/4 3/4

    but I am having trouble solving for the stationary distribution and expected number of trips. Somehow, I don't think the stationary distribution even exists. But it should. What am I missing? Thanks.
     
  2. jcsd
  3. May 12, 2009 #2
    hello carolyn

    your transition matrix is same as expected.And the stationary distribution exists. you can find the stationary distribution by solving it as two linear equations.
    let 'Pij' be the element, i=1,2 and j=1,2 (for the given transition matrix).

    P0=P0*P00+P1*P01; P1=P0*P10+P1*P11

    here P0 and P1 are unknowns , the elements of stationary distribution.
    Pij(P00,P01,P10,P11) are the transition matrix elements.

    By substituting you will get P0=P0*0+P1*(1/4); P1=P0*1+P1*(3/4).

    Also the condition to find P0 and P1 is P0+P1=1(the raw sum =1);

    then u will get a stationary distribution of
    [1/5 4/5]
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Markov chain - stationary distribution - please help
  1. Markov Chain (Replies: 1)

  2. Markov chains (Replies: 0)

  3. Markov Chain (Replies: 1)

Loading...