I have a fairly easy question but I did not manage to find the solution. I
want to regress a variable on another one but I do not want to give the same
importance to all my points. For example, let's immagine I have:
y | x
--------
1 | 2
2 | 4
3 | 6
To make a standard OLS, I would type reg y x.
Now, I want to give a much higher importance to the last point than to the
others. For example I want to weight the first two points by 1 and the last
by 1.4. This should be done, for example, because of a non representative
sample where an individual (or a group of individuals) has been clearly
under-represented.
Does anyone know how to do it?
Thank for your help
Reards
VIncenzo
_________________________________________________________________
Consigue aqu� las mejores y mas recientes ofertas de trabajo en Am�rica
Latina y USA: http://latam.msn.com/empleos/
