In our psychology experiment we have found Markov Chain transition probabilities to be a good way to analyse data, and the convergence of two Markov Chains on a random walk (we used 10,000 walks) to be a statistic that describes the data nicely (they converge around step 60).

I can't find how one could report this in a psychology paper though. Any help or nice examples? Many thanks!

  • $\begingroup$ Can you add more details about what you did and how you interpret the results? This does not sound to be related to what Markov chain Monte Carlo typically means. $\endgroup$ – Juho Kokkala Sep 17 '15 at 10:44
  • $\begingroup$ Oh, ok, it's very possible that we may have misunderstood something. So we created two probabilistic transition matrices on the basis of our data in two conditions. They were made on averages across the trials. Then we started random walks (10,000) from state 1 on each case and saw that by step 60 both of them had pretty much reached a stable position in state 3 (it was just a very strong attractor). But the time for reaching state 3 was different for them. I'd be trying to report that on the basis of the Markov matrices, both conditions would reach state 3 given enough time. Many thanks! $\endgroup$ – puslet88 Sep 17 '15 at 12:20
  • $\begingroup$ If you know of sources or nice examples I should read instead, they will also be very welcome! Thanks. $\endgroup$ – puslet88 Sep 17 '15 at 13:47
  • $\begingroup$ You should edit the question to contain information such as i) what kind of data you have, ii) how you have estimated the transition probabilities, iii) what are you trying to find out and why. $\endgroup$ – Juho Kokkala Sep 17 '15 at 14:30
  • $\begingroup$ Is this related to your other question stats.stackexchange.com/questions/172171 ? $\endgroup$ – Juho Kokkala Sep 17 '15 at 14:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.