I want to choose two-pairs for pairtrading. For pairtradings, two pairs need to pass two tests, the Johansen test (for cointegration) and the P.P. test (for stationary). As I knew, if they related in cointegration, they would be stationary. But, My data passed Johansen test, but they couldn't pass P.P. test.

Why does this result occur?

| cite | improve this question | | | | |
  • $\begingroup$ What do you mean by “pass” the PP test? $\endgroup$ – The Laconic Nov 6 '18 at 14:37

I'm not sure I understand what you are asking. I think you are asking why two stocks are co-integrated but don't pass a test for non-stationary by themselves. If a stock is a constant price, say $10, then it is co-integrated with EVERYTHING. So it is kind of pointless to use for pair trading. So maybe your test for stationarity is just finding that one or both of your stocks are basically constant with a random noise.

By the way, if you want to find pair to trade with, try PairFinder app, at http://www.kizbit.com (which directs to Google Android Play store).

| cite | improve this answer | | | | |

Your Answer

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

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