Dear Eric

I am a Ph student in South Korea and recently I am studying the QARDL-ECM model.  I read your blog about the introduction to the QARDL model recently, it is really helpful. And I also read the Jin Seo Cho's (2015) paper, "Quantile Cointegration in the Autoregressive
Distributed-Lag Modelling Framework"
. But I still do not know how to run the QARDL-ECM. When I run the QARDL, I only can get some partial results(Long-run Parameter Estimates (β), Short-run Parameter Estimates (Φ), Short-run Parameter Estimates (γ)) as you wrote in the QARDL model blog. There's no constatn term(α) and the adjustment coefficient(ρ) in the results obtained by the QARDL model.

So my question is that if I want to get QARDL-ECM results like Jin Seo Cho's (2015) paper:

1) Is it correct to follow the sequence of the following thress steps: Do the ECM first, and then get the ARDL-ECM, and final do the Quantile regression in the ARDL-ECM.

2) If it is wrong, would you mind to tell me the correct sequence to do the QARDL-ECM in the right way please.

I really need your help and look forward to your reply please.

Thank you very much.

1 Answer



From my interpretation of the paper, yes it does seem that you are correct in your proposed sequence.

The authors first estimate the ARDL-ECM model (using OLS) as a baseline case and then estimate the QARDL-ECM model following that.


You must login to post answers.

Have a Specific Question?

Get a real answer from a real person

Need Support?

Get help from our friendly experts.

Try GAUSS for 14 days for FREE

See what GAUSS can do for your data

© Aptech Systems, Inc. All rights reserved.

Privacy Policy