Re: [Gretl-users] Exact collinearity and modprint
Thanks a lot.
Giuseppe
To Allin:
The regression is actually just a FE panel data model, but it follows a
concentrated NLS to estimate a Panel Smooth Transition Regression model.
So, I would like to have more control on the output (s.e. and
statistics).
On Thu, 2011-11-03 at 15:14 +0100, Ignacio Diaz-Emparanza wrote:
El 03/11/11 10:31, Giuseppe Vittucci escribió:
> Dear all,
>
> when in a regression one or more variables are omitted due to exact
> collinearity, the command modprint leads to misleading results cause the
> initial list of regressors does not match the vector of coefficients
> returned by gretl.
>
> Is there a way to retrieve the list of regressors dropped by gretl or
> the list of regressors eventually used in the regression?
>
> Thanks
> Giuseppe
yes. This is automatically saved in $xlist you may call it after the
estimation command by means of for example:
list listname = $xlist