0
$\begingroup$

When I run a regression, all variables are I(1), the optimal lag according to SIC is one, and means I should do VECM (1-1=0) the coefficient of the error correction term (ECT) is negative but not significant. My question is:

  1. can I do VAR instead of VECM,
  2. Should I do VAR of differencs: means in E-views: D(dependent Variable) D(Variable1) D(Variable2) D(Variable3)..and so on. Or VAR (-1) means VAR one lag.
Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

it depends on : first: look at the rank in the johansen co-integration test, if it is full, not full or null. if the rank is full ---> run VAR. If the rank not full ---> vecm. If the rank in null ---> defferenciate and then run VAR. Michel Johansen. Sun University.

Improve this answer
$\endgroup$

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.