According to Wikipedia the Johansen cointegration test "is a procedure for testing cointegration of several, say k, I(1) time series". Link

Now, if all my variables are I(0), can I still test co-integration with Johansen? Or: How can I potentially test co-integration?

I found Granger causality with a Wald test of my time series and would like to use the cointegration test to double-check whether I find cointegration where I also found causality (this should be the case).


1 Answer 1


Consider the definition of cointegration: There exists a linear combination of several time series that is stationary despite the fact that each of the individual series is nonstationary.

Hence, per definition, cointegration can only occur if the individual series are nonstationary.

To check for a relationship among stationary time series, you can, for example, use a standard t-test (possibly using robust standard errors to account for serial correlation).


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.