Hi,

As far as I can see, Allin, Riccardo (Jack) and Sven prefer to let the EG test as it is. I don't have any problem with that, of course. But (there is always a "but") I still do not fully agree with you:

1.- I just made some simulations with the EG test (50,000 replications). It turns out that, as Sven said, running the EG test without including the trend in the cointegrated equation (when the variables are independent I(1) plus drift) and then doing the DF on the residuals, slightly modifies the critical values. Curiously enough, including the trend in the cointegrated equation also modifies the critical values of the DF test on the residuals (i.e., according to my simulation, both sets of simulated CV are different to the EG critical values). So, both "situations" are using incorrect critical values. The real solution is to provide new critical values.

2.- I agree with Riccardo (Jack) that if the outcome of the test changes when a few observations are removed, then the test is not quite reliable. But that was not my argument. If you have 120 obs, why should we only use 90 obs. to draw inference? Sven said that getting the same sample sizes for all the variables would make possible a comparison between the ADF tests. I would rather say that the main point of doing ADF tests to the observed variables is to be sure that the variables are I(1), and not to compare the tests. It's always better to draw inference using all the available information.

3.- Allin said the test cannot be considered as rocket science and I agree. That said, it remains an option in Gretl, next to the Johansen test. The latter allows the user to include/remove deterministic elements in the cointegrating vector. Why would that not be possible with the EG?

Maybe all this is not the hottest topic on earth, so I will not insist anymore. But (!) it would be nice to have more flexibility when doing the EG test,

Friendly,

Daniel

From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: Gretl list <gretl-users@lists.wfu.edu>
Sent: Tuesday, May 8, 2012 3:52 PM
Subject: Re: [Gretl-users] issues with the DF/EG tests

On Tue, 8 May 2012, Allin Cottrell wrote:

> Count me as one such! The Engle-Granger test regressions are
> not rocket science and if one wants to do something
> non-standard that's fine, but there's little point in larding
> the "coint" command with additional options.

On a more general note: all inference in the unit root area is asymptotic. Actually, it's VERY asymptotic; in some cases, before p-values can be trusted at literal value, you may need to use thousands of observations, if not more. If your results change dramatically after dropping a few observations, I personally wouldn't trust too much either the full-sample results nor the reduced-sample ones.

--------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Economia

Università Politecnica delle Marche
(formerly known as Università di Ancona)

r.lucchetti@univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
--------------------------------------------------
_______________________________________________
Gretl-users mailing list
Gretl-users@lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users