I have constructed a mean reverting spread using two indexes. I know they have to be mean reverting, but when plotted side by side they are mean reverting for a little bit and then deviate and head one direction, all because of a few returns. I was wondering what the proper way of normalizing the trend or dealing with a non-stationary mean in a reverting series?

  • $\begingroup$ I think I got it. Detrending the data has seemed to be the best way to normalize it. $\endgroup$ – user5003 Mar 25 '13 at 4:35
  • 2
    $\begingroup$ Can you reply to your own question giving more details about the steps you used to do it ? Or was it as simple as estimating the constant parameter and removing it ? $\endgroup$ – BlueTrin Mar 25 '13 at 10:39

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.