Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: About weighted least squares for grouped data

From   Lina C <>
Subject   st: About weighted least squares for grouped data
Date   Fri, 8 Jul 2011 14:05:22 +0100

Dear all,

I'm trying to run a weighted least squares regression, weighting by
the variance of the error term using grouped data. I found that there
is a program wls0 that run this procedure, and it has an option w( .)
to indicate the weighting variable. I was wondering if in my case I
need to estimate first the variance of the residual of this regression
and then run the wls0 procedure.

Yj = b0*Xj + ej ; where j indicates the group.

Thank you.

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index