Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: How to get standard errors for a generated regressor

From   "Stas Kolenikov" <>
Subject   Re: st: How to get standard errors for a generated regressor
Date   Thu, 8 Mar 2007 09:29:27 -0600

Hey Kristin,

you could consider looking up literature on Murphy-Topel standard
errors -- the original paper is treating the linear regression case
only though, but there might be some extensions of it available
around; also there's a couple of SJ papers on that stuff. Or, better,
you could consider moving to a full fledge structural equation
modeling framework where you estimate simultaneously the factor part
of the model and the regression part of the model. For the latter, you
would need to figure out -gllamm- package, see

On 3/8/07, Kristin J. Kleinjans <> wrote:
Dear Statalist,

I estimate both a multinomial and an ordered logit model, where two of
the independent variables have been derived using factor analysis (I
use the predicted factors). I was hoping that one of you might know how to
get the correct standard errors for these?

I use Stata 8 (intercooled).

Thank you and best wishes,

Kristin J. Kleinjans
University of Aarhus
School of Economics and Management
Building 322
8000 Aarhus C
Phone: +45 8942 1624

*   For searches and help try:

Stas Kolenikov
*   For searches and help try:

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