On Mon, February 6, 2006 13:30, john w wrote:
Hi,
I just tried "omit exogenous variables" in VAR.
I estimated VAR with const, trend and one exogenous variable.
I tried to omit just one of the three available variables on the list and I
get the message: "Exact or near collinearity encountered".
Why is that?
Also in Gretl output names of selected and omited variables are not the
same.
This is strange. Are you using the most recent version? Things appear to be
working ok here. Could you post your data or send them to me privately so I
can see what's going on?
Ps. VAR lag selection is now great.
Thanks.
--
Riccardo "Jack" Lucchetti
Dipartimento di Economia
FacoltĂ di Economia "G. FuĂ "
Ancona