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: FGNLS estimator

From   Jordan Silberman <>
Subject   st: FGNLS estimator
Date   Sat, 19 Jan 2013 00:13:22 -0500

Hi Statalist,

Question regarding the feasible generalized nonlinear least squares
(FGNLS) estimator used for the nlsur command. It's been recommended in
some cases that a linear version of this estimator (FGLS) be used to
address heteroskedasticity in some cases. I would like to instead use
the FGNLS. Can anyone tell me if the FGNLS, like the FGLS, is robust
to heteroskedasticity? I strongly suspect that it is, but I want to
make sure.

*   For searches and help try:

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