General to Specific Modelling Automation

[Mila]

I've found the StepLS routine very useful in General to Specific modelling. I'm, however, trying to run regressions, remove insignificant (end) lags whilst keeping all lags to the last significant lag (StepLS removes all insignificant lags). I wonder if this is feasible to be programmed into Eviews? Any hints or pointers in the right direction would be appreciated.
Thanks!

[EViews Gareth]

I'm not sure I follow your statement entirely, but aren't you describing the difference between forwards and backwards stepwise?

[EViews Gareth]

Glenn pointed out to me that you want to remove lags based upon lag order (i.e. higher order lags are removed before lower ones). My bad for not reading your post carefully enough.

Should be relatively straight forward to program. Just take a look at the manual stepwise example, but rather than removing a variable based upon t-stat, use your different criteria.

[Mila]

OK, thanks! I will give it a try.