VECM: Dummy in cointegrating relationship

[folmer]

Hi,

I am modelling a VECM, and suspect that there is a structural break in the level of the cointegrating relationship somewhere. I would like to add a dummy to account for this, but I don't see a way to do this. Of course, I do not want the dummy to be modelled as the other endogenous variables, but that is the only way I see of adding something to the cointegrating relationship. Is there a way to do this?

[dakila]

read this blog http://davegiles.blogspot.com/2011/05/c ... -hips.html

[folmer]

Hi, thanks for your reply but I'm not sure I understand it correctly. It seems to me that the author is adding exogenous dummies to account for a break in trend. I am more interested in allowing for a level shift in the cointegrating relationship itself though.