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

st: GLS estimation Stata vs PCGive

From   Linn Renée Naper <>
To   "" <>
Subject   st: GLS estimation Stata vs PCGive
Date   Wed, 4 Nov 2009 09:25:40 +0100

I've been running some GLS random effects estimations using
xtreg y x, re robust.

Part of the work I am doing is to reproduce some estimations done by someone else.

The thing is that the original results are estimated in GiveWin/PCGive.

Using the xtreg command I am not able to reproduce the results from PCGive exactly.

In PCGive I know there are two different choices: GLS (using OLS residuals) and GLS (using within/between).
If you choose the option "GLS (using within/between)" the results are identical with my xtreg, re robust.
However the results I am trying to reproduce have been estimated using the PCGive option "GLS (using OLS residuals)".

My question is if there is a similar option in Stata?

Thanks anyone

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index