My Math Forum Interpreting the limiting distribution of markov chains?
 User Name Remember Me? Password

 Advanced Statistics Advanced Probability and Statistics Math Forum

 June 14th, 2016, 03:29 PM #1 Member   Joined: Jun 2012 Posts: 64 Thanks: 0 Interpreting the limiting distribution of markov chains? Hi all, I have a few questions on the interpretation and application of the steady state or limiting distribution for a markov chain transition matrix. Let's call this steady state matrix X where X = [0.49 0.51] The states are unemployed and employed from left to right or state 1 and 2. To arrive at this matrix, I used a transition matrix, T, that represented the transition probabilities between the states of unemployment and employment within the year and used the condition that xT = x where x is the stationary distribution. (i.e. x1 = 0.49 and x2=0.51) My first question is to clarify whether X also represents a matrix within a year? From what I understand about the steady state, X, it means that in the long run, the probability of being unemployed is 0.49 within the year? Based on some of the papers I read, we can also use the steady state to represent the "long run proportion of time that the chain spends" in a state. So for the above...does it mean that in the long run, a person will spend half a year unemployed and little more than half a year employed? Also, I have read that I can use X to determine the "expected number of steps(time) until the chain revisits" a state e.g. E[time from state 1 to 1] = 1/x1 . So x1 would be the expected number of steps the chain revisits the unemployment state given the chain started in the unemployment state i.e. 1/0.49 = 2. Does this mean that if I'm unemployed, and leave that state into employment...I expect to be unemployed again after 2 years? Therefore, if apply the above for person's working career before retirement, say for 40 years...it means that the person will spend about 20 years unemployed and expects to be unemployed every 2 years? The paper I'm quoting from is from Karl Sigman.

 Tags chains, distribution, interpreting, limiting, markov

 Thread Tools Display Modes Linear Mode

 Similar Threads Thread Thread Starter Forum Replies Last Post calypso Advanced Statistics 5 May 27th, 2016 01:32 AM interestedinmaths Advanced Statistics 5 February 3rd, 2015 03:29 AM lakshwee0292 Algebra 0 December 3rd, 2013 12:37 AM inequality Algebra 11 November 26th, 2009 03:12 AM aldors Algebra 0 December 31st, 1969 04:00 PM

 Contact - Home - Forums - Cryptocurrency Forum - Top