Interpretation of Johansen TEST

[diego45]

Hi could please tell me if what i say is good.




So here we can see that for 5% the trace statistic is superior to the critical value. So we reject the null hypothesis that "none" variable are co-integrated.
After we also rejet the null hypothesis that "At most 1" variable is co-integrated so there is not at most 1 co-integrated variable in the model.

So there is no cointegration right ?

I mean it's strange cause rejected "none" means that their is co-integrated variable ? if i reject that there are not co-integrated variable that means there at least one ? nop ?


thanks

[diego45]

edit

[aacasiobuk]

You have cointegration, with more than 1 cointegration vector. You cannot confirm that its two but you can assume that you have more than 1.