My Math Forum  

Go Back   My Math Forum > College Math Forum > Advanced Statistics

Advanced Statistics Advanced Probability and Statistics Math Forum

LinkBack Thread Tools Display Modes
June 14th, 2016, 03:29 PM   #1
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 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.
Keroro is offline  

  My Math Forum > College Math Forum > Advanced Statistics

chains, distribution, interpreting, limiting, markov

Thread Tools
Display Modes

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

Copyright © 2019 My Math Forum. All rights reserved.