VECM with Long-Run Restriction

[macro_man]

All:

Can you point me to any examples where a VECM is estimated with long-run BQ restrictions? Thanks.

[TomDoan]

See

http://www.estima.com/forum/viewtopic.php?f=4&t=730

and

http://www.estima.com/forum/viewtopic.php?f=4&t=263

[macro_man]

Thanks Tom. Two follow-up questions.

First, In the first link it is noted that with two cointegrating vectors this model would be easier to analyze in CATS. Does mean imposing the Blanchard-Quah restrictions(or other structural restrictions) is also done in CATS?

Second, is it possible that any tension could arises between imposing the BQ LR restrictions and the LR equilibrium restrictions imposed by the VECM? That is, is it possible that the two types of LR restrictions could be imposing structure on the data ocntradict eachother?

[TomDoan]

Thanks Tom. Two follow-up questions.

First, In the first link it is noted that with two cointegrating vectors this model would be easier to analyze in CATS. Does mean imposing the Blanchard-Quah restrictions(or other structural restrictions) is also done in CATS?

No. That was a response to a question which didn't relate directly to the original post.

Second, is it possible that any tension could arises between imposing the BQ LR restrictions and the LR equilibrium restrictions imposed by the VECM? That is, is it possible that the two types of LR restrictions could be imposing structure on the data ocntradict eachother?

The second link has some discussion about that. It's just that some LR zero restrictions will no longer help to identify the (SVAR) model because they are forced by the combination of cointegration and other LR zero restrictions.