Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Is it okay to feed $I(0)$ variables into the Johansen procedure? I've read three sources that seem to state that this is not what you're supposed to do. However, whenever I've done this, I notice that $\Pi$ is full rank and so it leads me to a VAR and therefore I don't see any problem with this.

share|improve this question

Having I(0) variables is not a theoretical issue. Please see pp 5-6 Hjalmarsson and Österholm (2007) on this.

share|improve this answer

This is an expected result. The matrix $\Pi$ has full rank, when the process is stationary. Of course Johansen procedure usually requires that the time series should be checked for unit roots first. The null hypothesis is that time series are unit-roots and they are cointegrated. If your variables are $I(0)$ then the first step should eliminate the need to use Johansen's test. I would hesitate to use Johansen's test for testing whether the processes are $I(0)$, since it was not designed to be used as such.

share|improve this answer

Don't deliberately difference the data to induce Stationarity. While using Johansen Trace Test to test for co-integration ALWAYS take the the data in levels.Then the number of co-integrating vectors obtained from Pi Matrix will give correct results

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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