El Miércoles, 26 de Octubre de 2005 09:10, maite escribió:
Sorry, but I am still having trouble restricting the sample within a
Monte
Carlo loop. I had already tried the loop --progressive option and it had
not worked. I tried again as suggested and still I only get one observation
for btilda. Any more suggestions ?
Maite,
you must have into account that you are creating a new dataset in a new file.
So, for working with the 1000 obs. of bhat you saved, you must close your
current dataset and open the file test.gdt you created. I run the program
----
nulldata 50
genr x=100*uniform()
loop 1000 --progressive
genr eps=3*normal()
genr y=5+2*x+eps
smpl (obs<26) --restrict
ols y const x
genr bhat=coeff(x)
store test.gdt bhat
endloop
---
and it worked for me (gretl version 1.4.1 in Mandrivalinux 2005). If it is not
working in your case, Allin will need to know your gretl version and
operative system (Windows 98, XP, Linux ...) for trying to solve the problem.
--
Ignacio Díaz-Emparanza
Dpto. de Economía Aplicada III (Econometría y Estadística)
UPV-EHU
http://www.bl.ehu.es/~etpdihei/