Is there ever a reason to use a first differenced VAR over a VECM when all your variables are I(1) and co integration exists?
The reason why I ask is because I see in the most recent Bank of Canada Analytic note on the minimum wage increase (page 15) it seems that opt for a reduced form VAR over a VECM.
Why is this method used?
Source:http://www.bankofcanada.ca/wp-content/uploads/2017/12/san2017-26.pdf
when all the variables in the model are integrated at order one, it apt to use cointegration. If the results diclose the presence of cointegration, we should go for Error Correction model.
VAR does not serve the purpose of VECM. So, VAR is not a substitute for VECM
You should consult the authors; however, a researcher might elect to use a VAR model instead of a VECM model if identification is more important to them than long run coontegration effects. This is true even though we understand the VECM is more appropriate.
When the variables are cointegrated, it is more appropriate to use a VECM. On the other hand, it is also possible to use a difference VAR model according to the desired objective. In this case, for example, authors are not interested in short-term and long-term relationships.
- Badges
- Science topic
- Similar topics
- Biological Science
- Anatomy
- Tissue