Page 1 of 1

ARDL Model Estimation Problem

Posted: Fri Apr 19, 2019 7:03 am
by Zhenyang226
Dear all,

I have a question regarding the ARDL model. I am estimating the ARDL with the unrestricted constant and unrestricted trend, I could only find the short-run constant term, but couldn't find the long-run constant term (Please kindly check the attachment file for my long-run estimation). I am new to using Eviews and haven't been able to find any related problem online. Hope anyone of you could answer my question. Thank you.
Screenshot 2019-04-19 at 08.59.30.png
Long-run estimation
Screenshot 2019-04-19 at 08.59.30.png (15.53 KiB) Viewed 515 times

Re: ARDL Model Estimation Problem

Posted: Fri Apr 19, 2019 11:47 am
by EViews Mirza
Hi!

The reason that you don't see the constant or trend terms in the long-run levels equation is due to the fact that you are estimating ARDL using the deterministic scheme 5 which does not subject the constant and the trend to the long-run restriction. In other words, the constant and trend are NOT part of the long-run or cointegrating equation. They are however treated as short-term variables and are therefore treated as such in the numerous other auxiliary regressions presented in the ARDL output.

For further details, I would suggest you read our 3-part ARDL blog, but if you need a quick summary, refer to the last section in Part 1. The link is provided here: http://blog.eviews.com/2017/04/autoregressive-distributed-lag-ardl.html