cointegration with different integrated order

[happyjoy89]

hi

If there is a mix of I(0) and I(1) variables, can I still proceed Johansen cointegration test?

If I can, how do I set up the model for performing cointegration tets?

[Pandelis]

if you have more than 1 I(1) variables then theoretically they may be still co-integrated. If not then you use the unique I(1) variable in first differences, thus you have a stationary VAR models and you don;t have to test for co-integration.