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

Re: st: Can you get BETA coefficients for XTMIXED models?

From   Maarten buis <>
Subject   Re: st: Can you get BETA coefficients for XTMIXED models?
Date   Thu, 1 May 2008 15:10:58 +0100 (BST)

--- James Laurence <> wrote:
> I was wondering if it is possible to get BETA coefficients when using
> an XTMIXED model/command?
> I've looked in POSTESTIMATION and tried LISTCOEF but none of these
> seem to work.

Sometimes standardized regression coefficients, i.e. the regression
coefficients of a regression of a standerdized dependent variable on
standardized independent variables. With a multilevel model this leads
to a nontrivial question: do you standardize within groups or across
groups. Say you have data on many countries, do you divide the
variables by their national standard deviation, or by the overall
standard deviation? I personaly tend to prefer the overall standard
deviation, but I do see the arguments in favor of standardizing using
national standard deviations.

Notice that standardizing a variable often means also subtracting the
mean. If you subtract the national mean, you in effect do a procedure
called demeaning, which leads to a fixed effects model. So you might
not want to do that.

Hope this helps,

Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

Sent from Yahoo! Mail.
A Smarter Email
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index