Re: [Gretl-devel] coint2 with exogenous variables

On 11/05/2011 09:35 PM, Allin Cottrell wrote: > On Sat, 5 Nov 2011, Andreas Noack Jensen wrote: > >> Den 04/11/2011 kl. 18.16 skrev Allin Cottrell: >> >>> So I'll try to track down >>> the source of the difference with Tr76 added. >> >> I have looked at the problem again and found the error. It is an >> easy fix, I guess. The restricted variables are lagged one period. >> For that reason the sample is also changing if the restricted >> variable is excluded. The attached modified script gives exactly >> the same results in Gretl and Ox. > > Yes, I came to the same conclusion, and I've just committed a change > to gretl CVS that renders the results the same (from gretl and Ox) > for your original script. I'm not sure but could it be that we had discussed the question of whether the exogenous restricted variables should be lagged or not? Ultimately it's a matter of convention, of course, but what I may have argued back then is this: since the error correction term also (usually) appears with a t-1 index, I'd say it's not wrong to include x_{t-1} instead of x_t.

>> As a side note I have a small suggestion. It is unfortunate to >> test cointegration rank in the models with unrestricted >> deterministic terms so maybe they should not be there. If you >> really want to work with a model with unrestricted terms after a >> rank test procedure you can just make the exclusion restriction on >> he cointegration vectors and I think that is the "right" way to >> end up in that model. At least I shall suggest that the default is >> one the models with restricted terms. > > You're suggesting that we shouldn't allow including unrestricted > deterministic terms in cointegration testing (though we should allow > them in VECM specifications, having determined cointegrating rank)? > I think you must take into account that he has been trained in Copenhagen... They put a lot of emphasis on separating the cointegration rank determination from the specification of the deterministics. We can discuss this further if you like, but I am certainly strongly against disabling some of the possible test setups. Note that the most widespread case of an unrestricted constant also belongs to that category.