DOLS estimation and Dummy

[renudr]

Hi,

1) I have estimated a DOLS cointegrating equation between trade in services and trade in goods( 1 dep variable and 1 independant variable), I was wondering if in the estimation box I can just put my dummy variable to account for recessionary periods as an independant variable or is there another way to do this? To put it as an independant variable appears a bit strange to me as by theory all the independant variables in a DOLS are led and lagged and it doesn't make sense for that to happen with the dummy. Also, I am getting statistically insignificant results for the dummy coefficient when I do this. Is there another way to do this, or is this correct. Also, in your opinion, would not accounting for structural breaks in the DOLS model still make this model a credible one to test for cointegration?

2) Also, just to confirm, the output of the DOLS estimation reveal the long run relationship between the variables right( given that Engle Granger test on the error correction term shows a cointegrating relationship)? I performed an ECM model and the coefficient estimates on the differenced variables are not statistically significant whilst my DOLS coefficients are and was wondering if it makes sense to report the DOLS estimates as the long run relationship.

Thank you. Would appreciate a response soon! or at least point me to where I can find the answer...

Greatly appreciate your advice, thanks!