In the Johansen test, what if I am using a total of 5 variables and I reject the null hypothesis for rank 0 and rank 1, however ranks 2, 3 and 4 were not rejected because the P-value was greater than 5%. Do I then proceed to the VECM test with the assumption of cointegration existing between more than 0 to 1 variable?
No because there is no sufficient evidence to prove that there is cointegration between the variables . However, if the ranks 0, 1, and 2 null hypotheses were rejected you could have argued that there is cointegration between variables. This is because 3 out 5 ranks null hypotheses have been rejected ( majority caries the day ).
![ILLSLICK - KILLSHOT REMIX [FULL VERSION]](http://i.ytimg.com/vi/RIK8RrqIAzE/mqdefault.jpg)
In the Johansen test, what if I am using a total of 5 variables and I reject the null hypothesis for rank 0 and rank 1, however ranks 2, 3 and 4 were not rejected because the P-value was greater than 5%. Do I then proceed to the VECM test with the assumption of cointegration existing between more than 0 to 1 variable?
No because there is no sufficient evidence to prove that there is cointegration between the variables . However, if the ranks 0, 1, and 2 null hypotheses were rejected you could have argued that there is cointegration between variables. This is because 3 out 5 ranks null hypotheses have been rejected ( majority caries the day ).