[Gretl-users] problem with system instrumental variables

[Allin Cottrell]

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.

Allin.