EViews.com

Looking a little guidance to see if I am on the right track before I waste too much time trying this!

I have a model with 90 or so equations. I want to re-estimate the equations up to 2012 (as opposed to my default 2014). To do this I need to remove the dummy variable (D2013) from the equations that have it as a regressor. There are probably 20 or so that do have it (not all of them).

I am thinking I need to loop through all the equations and step through the @varlist to find "D2013" but I don't know if this is correct or how I would remove it if I found it was present. Any suggestions greatly appreciated!

Thanks in advance!

Assuming that each equation is specified by list and doesn't have ARMA/PDL terms, I think I would loop through the equations, use makegroup to make a group out of the variables in the equation, then remove d2013 from that group (setmaxerrs to high so it doesn't stop if it isn't in the group), then re-estimate the equation using that group.