I'm trying to account for structural breaks in the cointegrating relatioship by including breakpoint dummies interacting with the long-run coefficients (as well as the intercept shift term) using panel DOLS estimator. How can I tell EViews (9.5) not to include these dummy interactions in the short-run dynamics (not to have any lead/lag differences for them)? Even if I place them in the "deterministic regressors" EViews still add the differenced terms.
For technical questions regarding estimation of single equations, systems, VARs, Factor analysis and State Space Models in EViews. General econometric questions and advice should go in the Econometric Discussions forum.
2 posts • Page 1 of 1
Unfortunately, there is no way using our current engine. You can, of course, just estimate the DOLS equation directly by adding your leads, lags, etc. to a standard equation specification.
Who is online
Users browsing this forum: Bing [Bot] and 6 guests