[Gretl-users] problem with system instrumental variables

[Allin Cottrell]

On Thu, 7 Feb 2008, Allin Cottrell wrote:

> On Thu, 7 Feb 2008, nadaud at centre-cired.fr wrote:
> 
> > I would like to estimate a system of two equations with one 
> > different variable and one instrumented variable (income). So I 
> > use the syntax for tsls to specifiy regressors and instruments 
> > on the same line...
> 
> Yes, that method is supposed to work, but it seems to have got 
> broken at some point (assuming it ever really worked!).
> 
> I'm investigating this now and will report back before long.

Please try the new snapshot

http://ricardo.ecn.wfu.edu/pub/gretl/gretl_install.exe

The more obvious problems are fixed, I think.  I will do some 
checking against R.

Allin.