When preforming Johansen cointegration test for 2 time series (the simple case) you need to decide the lag you want to use. Doing the test for different lag levels returns different results: for some lag levels the null hypothesis can be rejected but for others it can't.

My question is what is the right method based on the input data to decide what lag I need to use?

  • $\begingroup$ I believe this belongs to StatSE $\endgroup$
    – SRKX
    Jan 22, 2012 at 14:13
  • $\begingroup$ In my opinion, this question fits our domain well, it's clearly in the domain in the domain of financial econmetrics. However I feel professionals should be able to figure this out themselves. On the other hand: a good, comprehensive answer would be a boon to this site. $\endgroup$
    – Bob Jansen
    Jan 22, 2012 at 14:49
  • $\begingroup$ @Bootvis: I agree it can apply, but there is a dedicated site for pure statistics questions (it doesn't involve anything about finance). If the question was more specific/advanced and especially focused on finance, then yes. Here it's not. I didn't downvote though, I voted for a migration. $\endgroup$
    – SRKX
    Jan 22, 2012 at 15:43
  • $\begingroup$ I would deeply appreciate if someone who has a good answer could answer my question. I think it can help people in this group because it is a common problem when constructing a cointegration system. $\endgroup$
    – Freewind
    Jan 24, 2012 at 15:29
  • $\begingroup$ This question has been answered at StatSE. $\endgroup$
    – J. Morris
    Sep 7, 2013 at 10:23

1 Answer 1


If you are using Spatial Econometrics toolbox in Matlab you could use the lrratio function which implements a sequence of such tests beginning at a maximum lag (specified by the user) down to a minimum lag (also specified by the user). (more info in http://fmwww.bc.edu/ec-p/software/matlab/mbook.pdf)


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.