Re: [Gretl-users] Forecasts out of the sample
On Fri, 30 Apr 2004, Ignacio Díaz-Emparanza wrote:
I think it would be interesting to have a new entry in the /Model
data/ menu that allows calculating some forecasts and confidence
intervals of Y for new values of X. ¿Could you think about it?
Yes, that sounds useful. At present one can do this using "genr", but
an automatic way of doing it would be helpful.
smpl 1 40
ols y 0 x1 x2
smpl 1 60
genr fcast = coeff(0) + coeff(x1)*x1 + coeff(x2)*x2
print y fcast -o
--
Allin Cottrell
Department of Economics
Wake Forest University, NC